Correlation functions between monopoles and instantons

合集下载

外源性磷输入改变农业河岸带湿地中土壤溶解有机碳的复杂性

外源性磷输入改变农业河岸带湿地中土壤溶解有机碳的复杂性

Exogenous phosphorus inputs alter complexity of soil-dissolved organic carbon in agricultural riparianwetlandsMeng Liu a ,Zhijian Zhang a ,⇑,Qiang He b ,Hang Wang a ,Xia Li a ,Jonathan Schoer caCollege of Natural Resource and Environmental Sciences,China Academy of West Region Development,ZheJiang University,Yuhangtang Avenue 866,HangZhou,ZheJiang Province 310058,China bDepartment of Civil and Environmental Engineering,University of Tennessee,Knoxville,TN 37996-2010,USA cDepartment of Chemistry,Valparaiso University,Valparaiso,IN 46383,USAh i g h l i g h t sExternal P input stimulated the production of sediment active C fractions.External P input decreased DOC humicity and increased its microbial-derived sources. Sediments with gradient P loading rate had a blue shift of fluorescence fingerprint. Spectra measurements were helpful for describing sediment DOC composition.a r t i c l e i n f o Article history:Received 29June 2013Received in revised form 23September 2013Accepted 25September 2013Available online 30October 2013Keywords:Dissolved organic carbon (DOC)Phosphorous (P)Riparian wetlandsStructural compositiona b s t r a c tHigh-strengthened farmland fertilization leads to mass inputs of nutrients and elements to agricultural riparian wetlands.The dissolved organic carbon (DOC)of such wetland sediments is an important inter-mediate in global carbon (C)cycling due to its role in connecting soil C pools with atmospheric CO 2.But the impact of phosphorus (P)on sediment DOC is still largely unknown,despite increasing investigations to emphasize P interception by riparian wetlands.Here,we simulated the temporal influences of exoge-nous P on sediment DOC of riparian wetlands by integrating gradient P loading at rates of 0%,5%,10%,20%,30%,and 60%relative to the initial total phosphorus content of the sediment with the purpose of illustrating the role of external P on the complexity of soil DOC in terms of its amount and composition.After incubating for nine months,a dramatic linear correlation between Olsen-P and fluorescent and ultraviolet spectral indices considered DOC skeleton was observed.Together with a more microbial-derived origin of DOC and a reduction of DOC aromaticity or humicity,the excitation-emission matrix had shown a blue shift reflecting a trend towards a simpler molecular structure of sediment DOC after P addition.Meanwhile,the content of soil DOC and its ratio with total organic carbon (TOC)were also increased by P loading,coupled with enhanced values of highly labile organic carbon and two C-related enzymes.While TOC and recalcitrant organic carbon decreased significantly.Such implications of DOC amounts and composition stimulated by external P loading may enhance its bioavailability,thereby inducing an accelerated effect on soil C cycling and a potential C loss in response to global climate change.Ó2013Elsevier Ltd.All rights reserved.1.IntroductionSince the 1980s,as agricultural activity has intensified signifi-cantly in Eastern China,the extensive application of fertilizers to farmlands and production of livestock manure have led to great phosphorus (P)loss from agricultural areas to adjacent ecosystems (Zhang and Shan,2008).Enhanced levels of P not only have led to great eutrophication with characteristic algal blooms (Roberts et al.,2012),but have also impacted the ecological remediationand resilience of such aquatic ecosystems (Jeppesen et al.,2005;Wang et al.,2013).For this particular purpose,many riparian wet-lands have been arranged in agricultural catchments in use world-wide to reduce the concentration of nutrients in through-flowing water and improve water quality (Verhoeven et al.,2006;Hoff-mann et al.,2009).Currently,total phosphorus (TP)content in riparian wetland sediment located in the southern region of the Taihu Basin has reached 169–1200mg kg À1after interception (Wang et al.,2010).Some researchers have found the decomposi-tion of longer-term or mineral-associated soil carbon (C)fractions and soil organic C sink strengths could be enhanced by P availability (Mack et al.,2004;Cleveland and Townsend,2006;0045-6535/$-see front matter Ó2013Elsevier Ltd.All rights reserved./10.1016/j.chemosphere.2013.09.117Corresponding author.Tel.:+8657186971854;fax:+8657186971719.E-mail addresses:zhangzhijian@ (Z.Zhang),qhe2@ (Q.He).Bradford et al.,2008).Other authors pointed out that undesired ef-fects such as additional risks for global warming may be induced as a result of nutrients overloading wetland sediment(Verhoeven et al.,2006;Zhang et al.,2012;Wang et al.,2013).However,the is-sue of sediment P accumulation on the sediment C pool has at-tracted less attention during past years,while policymakers mostly focus on the interception of P in riparian wetlands.Global warming significantly correlates to CO2loss from soils and creates a feedback to oceanic and land ecosystems(Cox et al.,2011).In soils,dissolved organic carbon(DOC),as the impor-tant fraction in the active soil organic C pool(Song et al.,2012), intimately correlates with CO2evolution from soil although DOC makes up only a small portion of total soil organic matter(Fang and Moncrieff,2005;Bengtson and Bengtsson,2007;Zhao et al., 2008).Meanwhile,soil DOC not only contains both substrates and end products of enzymatic reactions of varying molecular weight,but also constitutes the most bioavailable moiety for soil microorganisms(Song et al.,2012).Some researchers have also found that soil microbial respiration is significantly limited by large differences in the complexity of DOC(Fang and Moncrieff, 2005).Moreover,the microbial degradability of DOC and,there-fore,the relationship with C mineralization from soils may also be affected by its composition(Zhao et al.,2008),although not al-ways linearly dependent in amounts(Liu et al.,2012).The features of the DOC skeleton,particularly percent aromaticity,degree of structural conjugation,and humicity(Johnson et al.,2011;Guo et al.,2013),are of fundamental significance to indicating the retention or out-gassing processes for soil C pools(Wilson and Xenopoulos,2008).For instance,aromatic compounds probably derived from lignin are stable components,whereas carbohydrates are preferentially respired(Kalbitz et al.,2003).Thus,connecting DOC content variance with its structural complexity would better reflect the status and bioavailability of the soil C pool and its fur-ther retention.Among analytical characterization methods,the ra-pid,non-destructive,cost-effective,and informative density nature offluorescence spectroscopy(Guo et al.,2013)is well suited to provide informative data on the aromatic content and humicity of DOC,specific locations of differentfluorophores,andfluorescent characteristics of structure,functional group,configuration,heter-ogeneity,and molecular dynamics,which gives information about DOC composition(Fellman et al.,2008;Johnson et al.,2011;Guo et al.,2013).In soils,environmental factors such as climate(tem-perature,precipitation)and vegetation,or anthropic disturbances such as land-use,acidification,tillage,and application of fertilizer, may affect the DOC amount and composition indirectly via micro-bial consumption and lysis(Hishi et al.,2004;Jinbo et al.,2006). However,until recently,few studies have focused on the impacts of external P amendment on DOC of agricultural riparian wetland sediment,both in amount and composition,and thus the underly-ing mechanisms are largely unknown.Obviously,it is important to investigate the impact of P addition on wetland sediment DOC amounts,as well as its composition in agricultural areas.In this study,we designed a simulative experi-ment to illustrate the complex status of DOC by conducting labora-tory-scale incubation with P addition at rates of0%(P-0),5%(P-5), 10%(P-10),20%(P-20),30%(P-30),and60%(P-60)relative to the initial TP content of the sediment(0.29g kgÀ1;Supplementary Information,SI-1).Fluorescent and ultraviolet(UV)spectral mea-surements directly describe the effect on soil DOC composition. Moreover,the use of chemical or biochemical tests,with respect to their relationships withfluorescent and UV indices,gives in-sights into the complexity and biological degradability of soil DOC under disturbance of P inputs will be discussed.Further,we hypothesize that P loading could induce the variance of sediment DOC amount and composition to that of a simpler molecular struc-ture and enhance the bioavailability of soil DOC,thereby leading to a weakening effect on potential retention processes of soil C pool and accelerate soil C cycling in response to climate changes.2.Material and methods2.1.Site description and experimental designsRiparian wetland soil samples for this study were collected from the Southwest part of the Taihu Lake Basin(30°18051.8400N and119°54013.3700E).This area is one of the most productive and intensively farmed agricultural areas in the downstream delta re-gion of the Yangtze River in Southeast China.This region possesses a subtropical monsoon climate with an average summer tempera-ture of28°C and an average annual rainfall of1550mm.The most common agricultural land uses in this area include ricefields,veg-etable gardens,aquaculture,and swine farms.During the pastfive years,the annual soil P application rate has been30–85kg haÀ1-yrÀ1.In order to best profile the impact of exogenous P on soil DOC features,sediment was selected from a site(pH7.24;mois-ture57%)that contained a relatively low initial TP concentration of0.29g kgÀ1,compared to other similar wetland sediments in this region(0.17–1.20g kgÀ1of TP)(Wang et al.,2010).An average water depth around this particular sampling riparian wetland was1.4m with macrophyte plants on bank,and there was no arti-ficial channel existing nearby this natural isolated pond.The or-ganic C of this riparian wetland stored in the sediment and no other C inputs in this region,and the basic properties of the wet-land sediment were provided in Table2.The sediment samples were collected using a lab-made stain-less steel sampler at a depth of0–10cm from20different points. Samples were wiped clear of macro-particles and transported on ice to the lab within3h after collection.Plastic barrels(35cm diameterÂ35cm height)werefilled with8kg of mixed wet soil (5kg dry weight)to a depth of20cm.Water-soluble superphos-phate(CaP2H4O8)was chosen to be the external P in our study, which was recognized by Justus Liebig as P-related fertilizer and applied widely to agricultural production(Brunner,2010).Super-phosphate of different mass was dissolved with deionized water, then the mixed solution was added to those soil-filled barrels homogeneously with a spray without disturbing soil cores,of which TP concentration accounted for0%(P-0),5%(P-5),10% (P-10),20%(P-20),30%(P-30),and60%(P-60)relative to the initial sediment TP content.See supplementary information(SI-1)for the rationale for selecting the spiking levels of P in the sediment sam-ples.After several minutes,all the samples were covered with a 10cm layer of deionized water and were incubated in the labora-tory.To avoid the growth of aquatic plants and disturbance from other environmental factors,the barrels were placed in lab with dam-board to keep the samples in the dark at room temperature (20–25°C)for the nine months incubation.Deionized water was replenished every three months to maintain the liquid depth.Trip-licates of each sample were prepared.2.2.Soil samplingSoil samples for multifarious analyses were collected after incu-bating them for nine months.Each barrel was divided into four sub-barrels to minimize edge effects for grab-sampling.The soil samples for each barrel consisted of four composited2cm diame-terÂ10cm deep cores.50g fresh soil was collected from each bar-rel and then divided into two aliquots.One aliquot was stored at 4°C in the dark for microbial biomass and enzyme studies while the other was air-dried and ground to pass through a1mm mesh sieve for subsequent chemical analyses and spectral measurements.M.Liu et al./Chemosphere95(2014)572–5805732.3.Analytical methods2.3.1.Spectral measurementsBased on reported methods(Wilson and Xenopoulos,2008;Guo et al.,2013),the Solutions for UV andfluorescence measurements to determine DOC structure were prepared by water extraction of the air-dried and sieved(100mesh sieves)soil in a1:5w/w ratio to Milli-Q water,then shaking for4h at room temperature.Extracts were centrifuged at7000rpm(Hitachi Inc.,CR22G,Japan)for 10min at4°C andfiltered through a0.45l m membranefilter. The quantity of DOC in the extract was measured with a total or-ganic carbon(TOC)analyzer(Shimadzu Inc.,TOC-VCHP,Japan).Four UV andfluorescence–related indices were determined in this study:fluorescence index(FI),humification index(HI),fresh-ness index(b/a),and specific UV absorbance at280nm(E280).De-tails about the determination of these indices are provided in the supplementary information(SI-2).Briefly,FI,b/a,and HI were determined using afluorescence spectrophotometer(Hitachi Inc., F-4500,Japan).While E280was determined using a UV scanning spectrophotometer(Shimadzu Inc.,UV-2550,Japan).Milli-Q high-purity water was used as the reference for all measurements. Spectralfluorometric3D excitation-emission matrix(EEM)mea-surements were obtained for excitation wavelengths from200to 400nm and at emission wavelengths from300to600nm at 5nm increments as previously described(Wilson and Xenopoulos, 2008).Data analysis was then performed using an in-house pro-gram SigmaPlot12.0.2.3.2.Chemical and biochemical analysisMeasurements of sediment TOC,DOC,TP,and available P(pH 8.5,0.5mol LÀ1NaHCO3extractable P,i.e.,Olsen-P)were con-ducted according to standard methods of physicochemical analysis (ISSCAS,1978;Westerman,1990).Due to the susceptibility of or-ganic C to KMnO4oxidation,the contents of three fractions of labile organic components in soil samples,namely highly labile organic carbon(HLOC),mid-labile organic carbon(MLOC),and labile or-ganic carbon(LOC),were determined using33,167,and 333mmol LÀ1KMnO4,respectively(Loginow et al.,1987).Recalci-trant organic carbon(ROC)was calculated as the difference be-tween these three labile C forms and the TOC.Soil microbial biomass C(MBC)and P(MBP)were determined by the chloroform fumigation extraction method(Inubushi et al.,1991).Moist soil samples were split into two subsamples with one immediately ex-tracted with either0.5mol LÀ1K2SO4for MBC or0.5mol LÀ1 NaHCO3for MBP,while the other was fumigated with chloroform and then extracted.Following centrifugation,C and P concentra-tions of soil microbial biomass were calculated from the difference between the fumigated and non-fumigated soil samples.2.3.3.Enzyme analysisThree eco-enzymes,b-1,4-glucosidase(BG),cellobiohydrolase (CBH),and acid phosphatase(AP)(Sinsabaugh et al.,2009)were se-lected as indicators of microbial nutrient demand in the C and P cy-cles,respectively.Sample suspensions were prepared by using a vortex mixer for1min to homogenize1g(wet weight)of soil with 125mL of50mmol LÀ1sodium acetate buffer(pH6.0to match the mean soil pH of the environmental samples).Sample suspensions, buffer,references,and substrates(the substrate solutions for BG, CBH and AP are4-MUB-b-D-glucoside,4-MUB-b-D-cellobioside, and4-MUB-phosphate,respectively)were pipetted into96-well blankfluorescent plates(Corning Inc.,costar3603,USA)following the strict order and position on the well plate according to the work of Saiya-Cork(Saiya-Cork et al.,2002).The micro-plates were covered and incubated in the dark at20°C for4h.Then,10l L of 1.0mol LÀ1NaOH was added to each well to stop the reaction and increase thefluorescence of residual substrates.Finally,standard high-throughputfluorometric detection of the enzyme assays were carried out using a Bio-Tek Synergy HT microplate reader(Bio-Tek Inc.,Winooski,VT,USA)with365nm excitation and460nm emissionfilters(Saiya-Cork et al.,2002).Enzyme activities were calculated and expressed as nmol hÀ1gÀ1(Sinsab-augh et al.,2008).2.4.Statistical analysisData were tested for homogeneity of group variances using the Pearson test.Data were square root transformed if necessary.The data were analyzed by analysis of variance using SPSS16.0statis-tical software.For the enzymes analysis,each data point was char-acterized by a response ratio(RR)and the log of the response ratio (L RR).RR was calculated as the mean of the experimental samples divided by the mean of the control samples(sample P-0)to provide an index of response magnitudes,while L RR was calculated as the log10of RR.The use of L RR is preferred over the use of RR because it equally weighs the negative and positive responses and facili-tates statistical analysis(Marklein and Houlton,2012).Positive values of L RR represented an increase in enzyme activity relative to the control sample,whereas negative values indicated sup-pressed activity.Before analysis,each class was summarized by the weighted mean of L RR(LÃRR).3.Results3.1.Structural complexity of soil DOC under phosphorus loadingSpectral analysis showed that P input was an excellent factor for soil DOC composition of the riparian wetlands(Table1).FI in soil solutions was raised from P-0to P-60by a range of0.65%to a max-imum6.5%.Meanwhile,the ratio of b/a was observed to follow the same trend as FI,with an increment of21–64%.As to aromatic sub-strate,E280and HI were reduced in all P loading samples by a fac-tor of3–12%and12–47%,respectively.Moreover,we found that all spectral indices changed significantly with P availability.Briefly,FI and b/a correlated positively with Olsen-P content(Fig.1A and C; p<0.05),but the relationship with respect to HI was negative (Fig.1D;p<0.05),while there existed a less linear correlation with E280(Fig.1B).Two remarkablefluorophores in3D EEMs were revealed from the tested six treatments:one at Ex/Em280–320/ 400–450nm and another at Ex/Em210–240/395–450nm(Fig.2). Usually,the peak at longer wavelengths was recognized as the humic-likefluorescence peak C in the UV region,while the other was determined to be the humic-like peak A in the visible region (Klotzbücher et al.,2012).Compared to humic-likefluorophores reported in Fig.2,a clear blue shift about the positions of the fluorophores both in the visible and UV region was observed with the increment of P loading,displaying a trend towards shorterTable1Spectral parameters of sediment dissolved organic carbon(DOC)in the tested samples,namelyfluorescence index(FI),humification index(HI),UV absorption at 280nm(E280),and freshness index(b/a).Treatment FI E280b/a HIP-0 1.53±0.05b0.63±0.08ab0.91±0.07cd0.48±0.01a P-5 1.54±0.01b0.64±0.03a0.88±0.08d0.42±0.01b P-10 1.56±0.02ab0.61±0.03ab0.90±0.03cd0.40±0.03b P-20 1.59±0.03ab0.56±0.01ab 1.00±0.06c0.32±0.01c P-30 1.59±0.05ab0.55±0.03b 1.31±0.02b0.22±0.01d P-60 1.63±0.07a0.57±0.02ab 1.49±0.03a0.22±0.01d Values in parentheses are standard deviations.Different letters listed beside the data represent significant differences at p<0.05 (Duncan test,One-way ANOVA).574M.Liu et al./Chemosphere95(2014)572–580wavelengths(black line)relative to the maximum emission of the humic-like peak which moved towards to the left side of the emis-sion axis(Fig.2).3.2.Overall C-P features and enzyme activities under P loadingThe data calculated from the mean value of three tested soils for each sampling pot(Table2)showed that soil TOC content de-creased about12–18%relative to each sample without P amend-ment.Meanwhile,DOC content increased by a factor of0.7–25.5%,as did the ratio of DOC/TOC with the incremental addition of P(Fig.3C).Soil TP and Olsen-P were significantly increased with the rate of P application.Remarkable increases were also found for MBC and MBP by a rate of22–50%and5–54%,respectively.The ratio of HLOC/TOC followed the enhanced tendency of soil MBC compared to blank treatment(Fig.3A),while the ROC/TOC ratio was found to decrease with the rate of P addition(Fig.3B).As summarized in Table3,the activities of both soil BG and CBH were stimulated consistently with P amendment.The RR of BG ranged by a factor of31–62%;at the same time an increment of 14–19%was found for CBH.The positive weighted mean of L RR illustrated an enhancement effect on the activities of these two C-related enzymes induced by external P loading.P amendment depressed AP activity due to all weighted means of L RR about AP, which showed negative values reduced by16–43%,except for P-5.3.3.Pearson correlation for soil chemical or biochemical properties in relation to soil DOC characteristicsPearson correlation analysis showed that soil chemical or bio-chemical properties had a strongly significant correlation with1.Relationships between spectral parameters and Olsen-P after incubation.Shown are selected univariate linear regressions with the highestthe structure of soil DOC(Table4).Both TP and Olsen-P greatly promoted the change of DOC composition,considering its signifi-cantly positive relationship with b/a,FI,and HI values(p<0.05).Moreover,the DOC composition intimately correlated with soil C-related features.For example,FI and b/a correlated positively to soil DOC,MBC,HLOC/TOC,DOC/TOC,and C-relatedenzymes,but negatively to the ROC/TOC ratio (p <0.05),while HI positively correlated with the ROC/TOC ratio (p <0.01),but negatively to MBC,DOC,and C-related enzymes (p <0.05).4.Discussion4.1.Structural complexity of soil DOC under external P loading The characteristics of soil DOC composition are quite different from one another in gradient P input treatments.High FI and b /a values for P amendment samples represent a ‘‘first flush’’of predominant microbial-derived sources that builds up soil DOC (Table 1)with an intimate correlation with soil DOC content and DOC/TOC ratio (Table 4;p <0.05).This increase is consistent with standard interpretation of FI values,where higher values are representative of microbial decomposition of soil C and most labile C sources have already been microbially acquired (Wilson and Xenopoulos,2008;Johnson et al.,2011).Some authors have re-ported that P availability is one of the limiting factors for microbial growth (Ahn et al.,2007).Thus,the input of external P can be responsible for the enrichment of microbial-derived origins in soil DOC (Table 4;p <0.05),which is supported by the significant linear correlation between Olsen-P and FI,as well as b /a (Fig.1A and C;p <0.05),illustrating an enhanced function of autochthonous pro-duction of soil DOC of a simpler molecular structure as well as microorganism activities after gradient P irrigation.At the same time,the humicity of soil DOC intimately correlated with soil TP content (Table 4;p <0.05)and showed a negative linear relation-ship with available P (Fig.1D;p <0.05).HI,related to the degree of condensation or conjugation (Klotzbücher et al.,2012;GuoTable 3Eco-enzymes as indicators of microbial nutrient demand in the cycles of sediment carbon (C)and phosphorus (P)respectively:b -1,4-glucosidase,acid phosphate and cellobiohydrolase in the tested sediment.TreatmentAcid phosphatase b -1,4-glucosidase Cellobiohydrolase RRL ÃRR ±ClRR L ÃRR ±Cl RR L ÃRR ±Cl P-5 1.000.002±0.013a 1.090.036±0.008c 1.360.128±0.078b P-100.83À0.082±0.005b 1.310.116±0.047b 1.340.122±0.094b P-200.84À0.076±0.015b 1.570.196±0.031a 1.590.201±0.012a P-300.63À0.201±0.020c 1.660.220±0.037a 1.560.191±0.049a P-600.57À0.247±0.014d1.770.247±0.039a1.620.209±0.047aR,the response ratio;L ÃRR ,the weighted mean;Cl,95%confidence interval.Values in parentheses are standard deviations.Different letters listed beside the data represent significant differences at p <0.05(Duncan test,One-way ANOVA).et al.,2013),was significantly smaller in the sample with the high-est P loading rate(Table1),characterized by the lowest ROC/TOC ratio(Fig.3B),and showed a positive relationship with the refrac-tory moiety(Table4;p<0.01).The same tendency is deduced for E280from UV absorbance(Table1),but without much linear sig-nificance to Olsen-P(Fig.1B).The decreased HI and E280values (Table1)indicate that soil DOC contains less aromatic compounds as gradient rates of P loading increases,which further verifies selective removal of aromatic structures with a shift towards less aromatic precursors induced by external P.A clearer trend,considered qualitative information of soil DOC composition,induced by P loading is displayed by3D EEMs (Fig.2).Previous studies have demonstrated that the humic-like component or aromatic C content is negatively correlated with bio-degradation of DOC(Fellman et al.,2008;Hassouna et al.,2012), andfluorophores with long emission wavelengths are highly con-jugated and more aromatic in nature(Klotzbücher et al.,2012; Guo et al.,2013).Thus,as the black line shifted to the left side of the emission axis(Fig.2),the weakened humicfluorescence inten-sity indicated that addition of external P drives the transition of soil DOC composition to much more labile components with a sim-pler molecular structure and a lower degree of aromatic polycon-densation,which coincided with the data collected by spectral measurements(Table1).Since the molecular structure of organic material has long been thought to determine long-term decompo-sition rates in soil humic substances(Schmidt et al.,2011),the re-duced degree of DOC humicity and aromaticity(Table1and Fig.1B)combined with enhanced microbial-derived production of soil DOC(Table1and Fig.1A and C)may weaken the chemical stability and residence time of soil organic C components(Zhao et al.,2008),thereby accelerating cycling of the soil C pool.4.2.Is the transition of sediment DOC composition harmful for the retention of wetland sediment C pools under P loading?Previous studies have demonstrated the importance of soil DOC composition to soil microbial respiration(Fang and Moncrieff, 2005)and C mineralization(Zhao et al.,2008).As mentioned above,P input is a significant factor for the transition of soil DOC composition.Does this trend towards simpler structural complex-ity of soil DOC increase the possibility of the loss of soil C fractions and have a bad impact on C sequestration at the same time?In order to test this hypothesis,soil C-P features and enzymology were examined synchronously after incubation for a better under-standing of the complexity of soil DOC subjected to P loading and its implications for the retention of wetland sediment C pools.As P does not have a significant gaseous removal mechanism and therefore remains in the system to which it is added(Bostic et al.,2010),P loading to wetland systems results in a chemical gradient with P non-limiting conditions for microbial nutrient de-mand(Ahn et al.,2007)and mitigates the severity of P limitation to microbial biomass after enzymatic mineralization(Roberts et al., 2012).An obvious increase in content of MBC and MBP infive treatments was observed under P addition(Table2)indicating that microbial utilization of C and P might simply be stimulated by add-ing P due to the greater availability of labile organic components (Fig.3A and C),which generally implies enhanced availability of substrates for microbial growth(Lipson and Schmidt,2004).The increased microbial biomass positively correlated to DOC origins of microbial-derivation and negatively to soil humicity(Table4; p<0.05),better indicating the enhanced bioavailability of soil DOC after P incubation.Meanwhile,the response of enzyme activ-ities has provided insight into organic matter decomposition and soil Cfixing(Sinsabaugh et al.,2008).The corresponding activities of C-related enzymes(BG and CBH)were significantly stimulated after P amendment(Table3;L RR>0)due to increased DOC content (Table2and Fig.3C)since DOC contains both substrates and end products of enzymatic reactions of varying molecular weight(Bon-nett et al.,2006).In addition,enhanced DOC in soil results in en-riched substrate abundance for microbial metabolism and supports the synthesis of new enzymes(Song et al.,2012).BG and CBH are enzymes that contribute to the degradation of cellu-lose and other b-1,4glucans into glucose by deconstructing micro-bial cell walls and reducing macromolecules to soluble substrates for microbial assimilation(Sinsabaugh et al.,2008;Peoples and Koide,2012).Such mechanisms combined with the results of en-hanced microbial utilization of soil C(Table2)are highly respon-sive to changes in soil DOC composition.The Pearson analysis shown in Table4verifies that the activities of BG were closely linked to FI and b/a(p<0.05),but negatively correlated to HI and E280(p<0.05),illustrating the decomposition of soil organic mat-ter(Sinsabaugh et al.,2008)induced by P addition.Moreover,we found an average decrease of12–18%in TOC(Ta-ble2)and a reduced ROC/TOC ratio(Fig.3B)after P addition. Although this data changes little supported by the statistic analysis among the six treatments,but we can certainly predict a loss of sediment organic carbon since a more stable stage of the carbon pool appears due to the activities of microorganism during the long-term incubation.The lowest TOC content and ROC/TOC ratio in P-60,along with the highest MBC content(Table2),DOC/TOC, HLOC/DOC(Fig.3A and C),and enzymatic values(Table3),may as-sert a loss of organic resources and metabolized organic material at faster rates due to microorganism growth and activities stimulated by external P loading.Similar changes in the soil C pool subjected to P amendment are supported by Bradford et al.,who demon-strated enzyme-catalyzed depolymerization induced by P inputs would increase decomposition of soil C fractions that constituted the longer-term C pool(Bradford et al.,2008).The refractory C components comprising the C pool in the sediment were decom-posed during incubation(Fig.3B),when other interfering C-im-ports such as aquatic plants and animals were excluded,which are revealed to be positively correlated with soil humicity (p<0.01)and negatively related to b/a(Table4;p<0.05).Since humification has been defined as the conversion of fresh organic matter inputs into more stabilized substances(Kirkby et al., 2013),the reduction of the refractory moiety(Fig.3B)and TOC fractions(Table2)in our study undoubtedly caused a loss in the soil C pool,which asserts an intimate relationship with soil DOCTable4Pearson correlation coefficients between sediment chemical or biochemical proper-ties and sediment dissolved organic carbon(DOC)spectral characters responding toexogenous phosphorus(P)application of the riparian wetland.FI E280b/a HITOCÀ0.3950.318À0.802**0.804**DOC0.554*À0.4270.847**À0.830**MBC0.542*À0.485*0.551*À0.731**HLOC/TOC0.395À0.1790.717**À0.793**ROC/TOCÀ0.542*0.322À0.674**0.581*DOC/TOC0.514*À0.3900.902**À0.881**TP0.708**À0.3850.874**À0.824**Olsen-P0.580*À0.3910.864**À0.753**MBP0.736**À0.490*0.894**À0.805**APÀ0.628**0.529*À0.918**0.933**BG0.661**À0.632**0.819**À0.946**CBH0.478*À0.3720.545*À0.783**Sediment chemical or biochemical properties and DOC spectral characters aredimensionless.Pearson correlation coefficients showed by correlation matrix.Thepositive data denotes positive correlation coefficient,while the negative one rep-resents the opposite correlation coefficient;the correlation increases with theabsolute numerical value.*p<0.05.**p<0.01.578M.Liu et al./Chemosphere95(2014)572–580。

Correlation

Correlation

CorrelationXu JiajinNational Research Center for Foreign Language Education Beijing Foreign Studies University2Key points•Why correlation?•What is correlation analysis about?•How to make a correlation analysis?–Case studiesWhy Correlation?4Three things that stats can do •1.Summarizing univariate data •2.Testing the significance of differences •3.Exploring relationships b/t variables5Three things that stats can do •1.Summarizing univariate data •2.Testing the significance of differences •3.Exploring relationships b/t variables6探究事物之间的关联•植物的生长是否浇水的多少有关系,有多大关系•足球成绩好坏是否与身体(体质、人种)有关?•兴趣高、成绩好•元认知策略使用越多,学习进步越快•学好统计学有利于身体健康Key ides of correlationanalysis8•Correlation: co ‐relation . The co ‐relation is represented by a ‘correlation coefficient , r .•The range of the coefficient: ‐1to 1.•Three critical values: ‐1, 0and 1.Strength of correlationPositive correlation Strength of correlation Direction of correlationDirection of correlation Positive correlationDirection of correlation•Less Negative correlation12Two main types of correlation•Pearson : standard type, suitable for interval data (e.g. score, freq.)•Pearson r coefficient•Spearman : suitable for ordinal/rank data•Spearman rho coefficient13Significance•Similar to t ‐test and ANOVA statistics, the correlation coefficients need to be statistically significant.< .05Sig./P 值/alpha (α)值Coefficient of Determination r Ær2Æ% of variance explained15Coefficient of Determination •The squared correlation coefficient is called the coefficient of determination .•Multiplied by 100, this proportion of variance indicates the percentage of variance that is accounted for.•Correlation coefficients of .30 account for about 9% of the variance. Correlation of .70 explains about 49% of variance.Effect sizeCase Study 1Is connector use by Chinese EFL learners correlated with theirwriting quality?SPSS ProceduresAnalyze‐Correlate‐Bivariate1921Reporting correlations•In correlation tables/matrices •Embedded in textCorrelation tablesCorrelation tables(Dörnyei2007: 227)2324Embedded in text •As one would expect from the extensive literature documenting the benefits of intrinsic motivation, there was a significant positive correlation between overall GPA and intrinsic motivation (r = .34, p < .oo1).(Dörnyei 2007: 227)Practice: CET4 and CET6 Correlational analysisHomework英语成绩是否与语文成绩有相关性?28Wrap Up & Look Forward •Correlation coefficients provide a way to determine the strength & the direction of the relationship b/t two variables.•This index does not ... demonstrate a causal association b/t two variables.29Wrap Up & Look Forward •The coefficient of determination determines how much variance in one variable is explained by another variable.•Correlation coefficients are the precursors to the more sophisticated statistics involved in multiple regression (Urdan 2005: 87).30Thank you32。

详细介绍Spearman's correlation相关分析(英文)

详细介绍Spearman's correlation相关分析(英文)

The boxplot for Uranium is fairly consistent with one from a normal distribution; the median is fairly close to the centre of the box and the whiskers are of approximate equal length. The boxplot for TDS is slightly disturbing in that the median is close to the lower quartile and the lower whisker is shorter than the upper one, which would be suggesting positive skewness. Also there is an outlier and Pearson’s correlation is sensitive to these as well as skewness. Since we have some doubts over normality, we shall examine the skewness coefficients to see if there is further evidence to suggest whether either of the variables is skewed.
If your data does not meet the above assumptions then use Spearman’s rank correlation!
Monotonic function
To understand Spearman’s correlation it is necessary to know what a monotonic function is. A monotonic function is one that either never increases or never decreases as its independent variable increases. The following graphs illustrate monotonic functions:

decay of correlation 数学名词

decay of correlation 数学名词

decay of correlation 数学名词Decay of correlation(相关性的衰减)refers to the decrease in correlation between two variables as the distance between them increases. It is a mathematical concept used to quantify the relationship between two variables across different spatial or temporal distances.1. The decay of correlation between rainfall and crop yield was observed as the distance between the two fields increased.雨量与农作物产量之间的相关性随着两个田地之间的距离增加而减弱。

2. The study analyzed the decay of correlation between interest rates and stock market performance over a one-year timespan.该研究分析了利率和股市表现之间的相关性在一年的时间内是如何衰减的。

3. As the distance between two cities increased, thedecay of correlation between their population sizes became more noticeable.随着两个城市之间的距离增加,它们的人口规模之间的相关性衰减变得更加明显。

4. The researchers used statistical methods to determine the decay of correlation between air pollution andrespiratory diseases in different neighborhoods.研究人员使用统计方法来确定不同社区之间空气污染和呼吸道疾病之间的相关性衰减。

相关系数与共同方法偏差

相关系数与共同方法偏差

相关系数与共同方法偏差Correlation and Common Method BiasCorrelation is the process of measuring the relationship between two variables and determining if a linear relationship exists between them. Common method bias occurs when two measures are highly correlated due to the same method of measurement being used. This can lead to inflated correlations or spurious results.Correlation is most frequently used when studying the relationship between two variables. The degree of correlation can range from -1 to 1, with -1 indicating a perfect negative correlation and 1 indicating a perfect positive correlation. If the correlation coefficient is 0, it indicates that there is no linear relationship between the two variables.Common method bias occurs when two measurements are based on the same data collection method. For example, if two methods of measuring employee satisfaction are based on the same survey, the responses of employees will likely be highly correlated due to the same method being used. This can lead to inflated correlations or spurious results.To avoid common method bias, researchers should use different methods of data collection when measuring the samephenomena. For instance, when measuring employee satisfaction, a researcher could use an in-person interview and a survey, as well as observation or focus groups. By using different methods, researchers can better ensure that the observed correlations are the result of a true relationship between the variables, and not due to the use of the same method of data collection. In summary, correlation is a statistical measure of the relationship between two variables. Common method bias occurs when two measurements are based on the same data collection method, leading to inflated correlations or spurious results. To avoid this, researchers should use different methods of data collection when measuring the same phenomena.。

协方差

协方差

1
18.05 class 7, Covariance and Correlation, Spring 2014
2
2.2
Sums and integrals for computing covariance
Since covariance is defined as an expected value we compute it in the usual way as a sum or integral. Discrete case: If X and Y have joint pmf p(xi , yj ) then
n m
n
m
p(xi , yj )xi yj − µX µY .CovFra bibliotekX, Y ) =
i=1 j =1
p(xi , yj )(xi − µX )(yj − µY ) =
i=1 j =1
Continuous case: If X and Y have joint pdf f (x, y ) over range [a, b] × [c, d] then
18.05 class 7, Covariance and Correlation, Spring 2014
3
So Cov(XY ) = E (XY ) − µX µY =
5 1 −1= . 4 4 Next we redo the computation of Cov(X, Y ) using the properties of covariance. As usual, let Xi be the result of the ith flip, so Xi ∼ Bernoulli(0.5). We have X = X1 + X2 and Y = X2 + X3 . We know E (Xi ) = 1/2 and Var(Xi ) = 1/4. Therefore using Property 2 of covariance, we have Cov(X, Y ) = Cov(X1 +X2 , X2 +X3 ) = Cov(X1 , X2 )+Cov(X1 , X3 )+Cov(X2 , X2 )+Cov(X2 , X3 ). Since the different tosses are independent we know Cov(X1 , X2 ) = Cov(X1 , X3 ) = Cov(X2 , X3 ) = 0. Looking at the expression for Cov(X, Y ) there is only one non-zero term Cov(X, Y ) = Cov(X2 , X2 ) = Var(X2 ) = 1 . 4

Correlations between zeros and supersymmetry

Correlations between zeros and supersymmetry
Date : November 19, 2000. Research partially supported by NSF grants #DMS-9970625 ( rst author), #DMS-9800479 (second author), #DMS-0071358(third author). 1
CORRELATIONS BETWEEN ZEROS AND SUPERSYMMETRY
PAVEL BLEHER, BERNARD SHIFFMAN, AND STEVE ZELDITCH
Abstract. In our previous work BSZ2], we proved that the correlation functions for simultaneous zeros of random generalized polynomials have universal scaling limits and we gave explicit formulas for pair correlations in codimensions 1 and 2. The purpose of this paper is to compute these universal limits in all dimensions and codimensions. First, we use a supersymmetry method to express the n-point correlations as Berezin integrals. Then we use the Wick method to give a closed formula for the limit pair correlation function for the point case in all dimensions.

《概率论与数理统计》基本名词中英文对照表

《概率论与数理统计》基本名词中英文对照表

《概率论与数理统计》基本名词中英文比较表英文中文Probability theory 概率论mathematical statistics 数理统计deterministic phenomenon 确立性现象random phenomenon 随机现象sample space 样本空间random occurrence 随机事件fundamental event 基本领件certain event 必定事件impossible event 不行能事件random test 随机试验incompatible events 互不相容事件frequency 频次classical probabilistic model古典概型geometric probability 几何概率conditional probability 条件概率multiplication theorem 乘法定理Bayes's formula 贝叶斯公式Prior probability 先验概率Posterior probability 后验概率Independent events 互相独立事件Bernoulli trials 贝努利试验random variable 随机变量probability distribution 概率散布distribution function 散布函数discrete random variable 失散随机变量distribution law 散布律hypergeometric distribution 超几何散布random sampling model 随机抽样模型binomial distribution 二项散布Poisson distribution 泊松散布geometric distribution 几何散布probability density 概率密度continuous random variable 连续随机变量uniformly distribution 平均散布exponential distribution 指数散布numerical character 数字特点mathematical expectation 数学希望variance 方差moment 矩central moment 中心矩n-dimensional random variable n-维随机变量two-dimensional random variable 二维失散随机变量joint probability distribution 结合概率散布joint distribution law 结合散布律joint distribution function 结合散布函数boundary distribution law 边沿散布律boundary distribution function 边沿散布函数exponential distribution 二维指数散布continuous random variable 二维连续随机变量joint probability density 结合概率密度boundary probability density 边沿概率密度conditional distribution 条件散布conditional distribution law 条件散布律conditional probability density 条件概率密度covariance 协方差dependency coefficient 有关系数normal distribution 正态散布limit theorem 极限制理standard normal distribution 标准正态散布logarithmic normal distribution 对数正态散布covariance matrix 协方差矩阵central limit theorem 中心极限制理Chebyshev's inequality 切比雪夫不等式Bernoulli's law of large numbers 贝努利大数定律statistics 统计量simple random sample 简单随机样本sample distribution function 样本散布函数sample mean 样本均值sample variance 样本方差sample standard deviation 样本标准差sample covariance 样本协方差sample correlation coefficient 样真有关系数order statistics 次序统计量sample median 样本中位数sample fractiles 样本极差sampling distribution 抽样散布parameter estimation 参数预计estimator 预计量estimate value 预计值unbiased estimator 无偏预计unbiassedness 无偏性biased error 偏差mean square error 均方偏差relative efficient 相对有效性minimum variance 最小方差asymptotic unbiased estimator 渐近无偏预计量uniformly estimator 一致性预计量moment method of estimation 矩法预计maximum likelihood method of estimation极大似然预计法likelihood function 似然函数maximum likelihood estimator极大似然预计值interval estimation 区间预计hypothesis testing 假定查验statistical hypothesis 统计假定simple hypothesis简单假定composite hypothesis复合假定rejection region拒绝域acceptance domain接受域test statistics查验统计量linear regression analysis线性回归剖析。

Density Correlation Functions in Calogero Sutherland Models

Density Correlation Functions in Calogero Sutherland Models
k n i zi ,
giving (6)
f=
k
f {n1 , . . . nk }
i=1
Wni ≡
k
f {n1 , . . . nk }|n1 , . . . nk
where f {ni } are expansion coefficients. Since here we will be interested only in chiral states, all ni can be chosen positive. This is the “string” picture, where the state |{ni } is interpreted as a collection of strings with winding numbers ni . Defining string creation and annihilation operators |n ≡ a+ n |0 and n|0 >≡ an |n which obey the commutation relations [an , a+ m ] = nδm,n and interpreting the action of (5) on (6) in terms of an and a+ n , we get[12,13] H= 1 mL′2
CERN-TH.7243/94 USC-94/008 hep-th/9404192 April 1994
Density Correlation Functions in Calogero-Sutherland Models
arXiv:hep-th/9404192v3 4 May 1994
Joseph A. Minahan

correlation coefficient 相关系数

correlation coefficient 相关系数

Pearson product-moment correlation coefficient In statistics, the Pearson product-moment correlation coefficient (sometimes referred to as the PMCC , and typically denoted by r ) is a measure of the correlation (linear dependence) between two variables X and Y , giving a value between +1 and −1 inclusive. It is widely used in the sciences as a measure of the strength of linear dependence between two variables. It was developed by Karl Pearson from a similar but slightly different idea introduced by Francis Galton in the 1880s.[1] [2] The correlation coefficient is sometimes called "Pearson's r."Several sets of (x , y ) points, with the correlation coefficient of x and y for each set. Notethat the correlation reflects the noisiness and direction of a linear relationship (top row),but not the slope of that relationship (middle), nor many aspects of nonlinear relationships(bottom). N.B.: the figure in the center has a slope of 0 but in that case the correlationcoefficient is undefined because the variance of Yis zero.DefinitionPearson's correlation coefficientbetween two variables is defined as thecovariance of the two variables dividedby the product of their standarddeviations:The above formula defines the population correlation coefficient, commonly represented by the Greek letter ρ (rho).Substituting estimates of the covariances and variances based on a sample gives the sample correlation coefficient ,commonly denoted r:An equivalent expression gives the correlation coefficient as the mean of the products of the standard scores. Based on a sample of paired data (X i , Y i), the sample Pearson correlation coefficient iswhere , and are the standard score, sample mean, and sample standard deviation respectively.Mathematical propertiesThe absolute value of both the sample and population Pearson correlation coefficients are less than or equal to 1.Correlations equal to 1 or -1 correspond to data points lying exactly on a line (in the case of the sample correlation),or to a bivariate distribution entirely supported on a line (in the case of the population correlation). The Pearson correlation coefficient is symmetric: corr (X ,Y ) = corr (Y ,X ).A key mathematical property of the Pearson correlation coefficient is that it is invariant to separate changes in location and scale in the two variables. That is, we may transform X to a + bX and transform Y to c + dY , where a , b ,c , and d are constants, without changing the correlation coefficient (this fact holds for both the population and sample Pearson correlation coefficients). Note that more general linear transformations do change the correlation:see a later section for an application of this.The Pearson correlation can be expressed in terms of uncentered moments. Since μX = E(X ), σX 2 = E[(X − E(X ))2]= E(X 2) − E 2(X ) and likewise for Y , and sincethe correlation can also be written asAlternative formulae for the sample Pearson correlation coefficient are also available:The above formula conveniently suggests a single-pass algorithm for calculating sample correlations, but, depending on the numbers involved, it can sometimes be numerically unstable.InterpretationThe correlation coefficient ranges from −1 to 1. A value of 1 implies that a linear equation describes the relationship between X and Y perfectly, with all data points lying on a line for which Y increases as X increases. A value of −1implies that all data points lie on a line for which Y decreases as X increases. A value of 0 implies that there is no linear correlation between the variables.More generally, note that (X i − X )(Y i − Y ) is positive if and only if X i and Y i lie on the same side of their respective means. Thus the correlation coefficient is positive if X i and Y i tend to be simultaneously greater than, or simultaneously less than, their respective means. The correlation coefficient is negative if X i and Y i tend to lie on opposite sides of their respective means.Geometric interpretationRegression lines for y=g x (x) [red] and x=g y(y) [blue ]For uncentered data, the correlation coefficientcorresponds with the the cosine of the anglebetween both possible regression lines y=g x (x) andx=g y(y).For centered data (i.e., data which have beenshifted by the sample mean so as to have anaverage of zero), the correlation coefficient canalso be viewed as the cosine of the anglebetween the two vectors of samples drawn fromthe two random variables (see below).Some practitioners prefer an uncentered(non-Pearson-compliant) correlation coefficient.See the example below for a comparison.As an example, suppose five countries are found tohave gross national products of 1, 2, 3, 5, and 8billion dollars, respectively. Suppose these same five countries (in the same order) are found to have 11%, 12%,13%, 15%, and 18% poverty. Then let x and y be ordered 5-element vectors containing the above data: x = (1, 2, 3,5, 8) and y = (0.11, 0.12, 0.13, 0.15, 0.18).By the usual procedure for finding the angle between two vectors (see dot product), the uncentered correlationcoefficient is:Note that the above data were deliberately chosen to be perfectly correlated: y = 0.10 + 0.01 x . The Pearson correlation coefficient must therefore be exactly one. Centering the data (shifting x by E(x ) = 3.8 and y by E(y ) =0.138) yields x = (−2.8, −1.8, −0.8, 1.2, 4.2) and y = (−0.028, −0.018, −0.008, 0.012, 0.042), from whichas expected.Interpretation of the size of a correlation CorrelationNegative Positive None−0.09 to 0.00.0 to 0.09Small−0.3 to −0.10.1 to 0.3Medium−0.5 to −0.30.3 to 0.5Large −1.0 to −0.50.5 to 1.0Several authors [3] have offered guidelines for the interpretation of a correlation coefficient. Cohen (1988),[3] has observed, however, that all such criteria are in some ways arbitrary and should not be observed too strictly. The interpretation of a correlation coefficient depends on the context and purposes. A correlation of 0.9 may be very low if one is verifying a physical law using high-quality instruments, but may be regarded as very high in the social sciences where there may be a greater contribution from complicating factors.InferenceA graph showing the minimum value of Pearson's correlation coefficient that issignificantly different from zero at the 0.05 level, for a given sample size.Statistical inference based on Pearson'scorrelation coefficient often focuses on oneof the following two aims. One aim is to testthe null hypothesis that the true correlationcoefficient is ρ, based on the value of thesample correlation coefficient r . The otheraim is to construct a confidence intervalaround r that has a given probability ofcontaining ρ.Randomization approachesPermutation tests provide a direct approachto performing hypothesis tests andconstructing confidence intervals. Apermutation test for Pearson's correlationcoefficient involves the following two steps:(i) using the original paired data (x i , y i ),randomly redefine the pairs to create a newdata set (x i , y i ′), where the i ′ are a permutation of the set {1,...,n }. The permutation i ′ is selected randomly, with equal probabilities placed on all n ! possible permutations. This is equivalent to drawing the i ′ randomly "without replacement" from the set {1,..., n }. A closely-related and equally-justified (bootstrapping) approach is to separately draw the i and the i ′ "with replacement" from {1,..., n }; (ii) Construct a correlation coefficient r from the randomized data. To perform the permutation test, repeat (i) and (ii) a large number of times. The p-value for the permutation test is one minus the proportion of the r values generated in step (ii) that are larger than the Pearson correlation coefficient that was calculated from the original data. Here "larger" can mean either that the value is larger in magnitude, or larger in signed value, depending on whether a two-sided or one-sided test is desired.The bootstrap can be used to construct confidence intervals for Pearson's correlation coefficient. In the "non-parametric" bootstrap, n pairs (x i , y i ) are resampled "with replacement" from the observed set of n pairs, and the correlation coefficient r is calculated based on the resampled data. This process is repeated a large number of times,and the empirical distribution of the resampled r values are used to approximate the sampling distribution of the statistic. A 95% confidence interval for ρ can be defined as the interval spanning from the 2.5th to the 97.5th percentile of the resampled r values.Approaches based on mathematical approximationsFor approximately Gaussian data, the sampling distribution of Pearson's correlation coefficient approximately follows Student's t-distribution with degrees of freedom N − 2. Specifically, if the underlying variables have abivariate normal distribution, the variablehas a Student's t-distribution in the null case (zero correlation).[4] This also holds approximately even if the observed values are non-normal, provided sample sizes are not very small.[5] For constructing confidence intervals and performing power analyses, the inverse of this transformation is also needed:Alternatively, large sample approaches can be used.Early work on the distribution of the sample correlation coefficient was carried out by R. A. Fisher[6][7] and A. K. Gayen.[8] Another early paper[9] provides graphs and tables for general values of ρ, for small sample sizes, and discusses computational approaches.Fisher TransformationIn practice, confidence intervals and hypothesis tests relating to ρ are usually carried out using the Fisher transformation:If F(r) is the Fisher transformation of r, and n is the sample size, then F(r) approximately follows a normal distribution withand standard errorThus, a z-score isunder the null hypothesis of that , given the assumption that the sample pairs are independent and identically distributed and follow a bivariate normal distribution. Thus an approximate p-value can be obtained from a normal probability table. For example, if z = 2.2 is observed and a two-sided p-value is desired to test the null hypothesis that , the p-value is 2·Φ(−2.2) = 0.028, where Φ is the standard normal cumulative distribution function.Confidence IntervalsTo obtain a confidence interval for ρ, we first compute a confidence interval for F( ):The inverse Fisher transformation bring the interval back to the correlation scale.For example, suppose we observe r = 0.3 with a sample size of n=50, and we wish to obtain a 95% confidence interval for ρ. The transformed value is artanh(r) = 0.30952, so the confidence interval on the transformed scale is 0.30952 ± 1.96/√47, or (0.023624, 0.595415). Converting back to the correlation scale yields (0.024, 0.534).Pearson's correlation and least squares regression analysisThe square of the sample correlation coefficient, which is also known as the coefficient of determination, estimates the fraction of the variance in Y that is explained by X in a linear regression analysis. As a starting point, the total variation in the Yaround their average value can be decomposed as followsiwhere the are the fitted values from the regression analysis. This can be rearranged to giveThe two summands above are the fraction of variance in Y that is explained by X (right) and that is unexplained by X (left).Next, we apply a property of least square regression analysis, that the sample covariance between andis zero. Thus, the sample correlation coefficient between the observed and fitted response values in the regression can be writtenThusis the proportion of variance in Y explained by a linear function of X.Sensitivity to the data distributionExistenceThe population Pearson correlation coefficient is defined in terms of moments, and therefore exists for any bivariate probability distribution for which the population covariance is defined and the marginal population variances are defined and are non-zero. Some probability distributions such as the Cauchy distribution have undefined variance and hence ρ is not defined if X or Y follows such a distribution. In some practical applications, such as those involving data suspected to follow a heavy-tailed distribution, this is an important consideration. However, the existence of the correlation coefficient is usually not a concern; for instance, if the range of the distribution is bounded, ρ is always defined.Large sample propertiesIn the case of the bivariate normal distribution the population Pearson correlation coefficient characterizes the joint distribution as long as the marginal means and variances are known. For most other bivariate distributions this is not true. Nevertheless, the correlation coefficient is highly informative about the degree of linear dependence between two random quantities regardless of whether their joint distribution is normal[1] . The sample correlation coefficient is the maximum likelihood estimate of the population correlation coefficient for bivariate normal data, and is asymptotically unbiased and efficient, which roughly means that it is impossible to construct a more accurate estimate than the sample correlation coefficient if the data are normal and the sample size is moderate or large. For non-normal populations, the sample correlation coefficient remains approximately unbiased, but may not be efficient. The sample correlation coefficient is a consistent estimator of the population correlation coefficient as long as the sample means, variances, and covariance are consistent (which is guaranteed when the law of large numbers can be applied).RobustnessLike many commonly-used statistics, the sample statistic r is not robust[10] , so its value can be misleading if outliers are present[11][12] . Specifically, the PMCC is neither distributionally robust, nor outlier resistant[10] (see Robust statistics#Definition). Inspection of the scatterplot between X and Y will typically reveal a situation where lack of robustness might be an issue, and in such cases it may be advisable to use a robust measure of association. Note however that while most robust estimators of association measure statistical dependence in some way, they are generally not interpretable on the same scale as the Pearson correlation coefficient.Statistical inference for Pearson's correlation coefficient is sensitive to the data distribution. Exact tests, and asymptotic tests based on the Fisher transformation can be applied if the data are approximately normally distributed, but may be misleading otherwise. In some situations, the bootstrap can be applied to construct confidence intervals, and permutation tests can be applied to carry out hypothesis tests. These non-parametric approaches may give more meaningful results in some situations where bivariate normality does not hold. However the standard versions of these approaches rely on exchangeability of the data, meaning that there is no ordering or grouping of the data pairs being analyzed that might affect the behavior of the correlation estimate.A stratified analysis is one way to either accommodate a lack of bivariate normality, or to isolate the correlation resulting from one factor while controlling for another. If W represents cluster membership or another factor that it is desirable to control, we can stratify the data based on the value of W, then calculate a correlation coefficient within each stratum. The stratum-level estimates can then be combined to estimate the overall correlation while controlling for W.[13]Calculating a weighted correlationSuppose observations to be correlated have differing degrees of importance that can be expressed with a weight vector w. To calculate the correlation between vectors x and y with the weight vector w (all of length n),[14][15]•Weighted mean:•Weighted covariance•Weighted correlationRemoving correlationIt is always possible to remove the correlation between random variables with a linear transformation, even if the relationship between the variables is nonlinear. A presentation of this result for population distributions is given by Cox & Hinkley.[16]A corresponding result exists for sample correlations, in which the sample correlation is reduced to zero. Suppose a vector of n random variables is sampled m times. Let X be a matrix where is the j th variable of sample i. Letbe an m by m square matrix with every element 1. Then D is the data transformed so every random variable has zero mean, and T is the data transformed so all variables have zero mean and zero correlation with all other variables - the moment matrix of T will be the identity matrix. This has to be further divided by the standard deviation to get unit variance. The transformed variables will be uncorrelated, even though they may not be independent.where an exponent of -1/2 represents the matrix square root of the inverse of a matrix. The covariance matrix of T will be the identity matrix. If a new data sample x is a row vector of n elements, then the same transform can be applied to x to get the transformed vectors d and t:This decorrelation is related to Principal Components Analysis for multivariate data.Reflective correlationThe reflective correlation is a variant of Pearson's correlation in which the data are not centered around their mean values. The population reflective correlation isThe reflective correlation is symmetric, but it is not invariant under translation:The sample reflective correlation isThe weighted version of the sample reflective correlation isReferences[1]J. L. Rodgers and W. A. Nicewander. Thirteen ways to look at the correlation coefficient (/stable/2685263). TheAmerican Statistician, 42(1):59–66, February 1988.[2]Stigler, Stephen M. (1989). "Francis Galton's Account of the Invention of Correlation" (/stable/2245329). StatisticalScience4 (2): 73–79. doi:10.1214/ss/1177012580. .[3]Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.)[4]N.A Rahman, A Course in Theoretical Statistics; Charles Griffin and Company, 1968[5]Kendall, M.G., Stuart, A. (1973)The Advanced Theory of Statistics, Volume 2: Inference and Relationship, Griffin. ISBN 0852642156(Section 31.19)[6]Fisher, R.A. (1915). "Frequency distribution of the values of the correlation coefficient in samples from an indefinitely large population".Biometrika10 (4): 507–521. doi:10.1093/biomet/10.4.507.[7]Fisher, R.A. (1921). "On the probable error of a coefficient of correlation deduced from a small sample" (/2440/15169) (PDF). Metron1 (4): 3–32. . Retrieved 2009-03-25.[8]Gayen, A.K. (1951). "The frequency distribution of the product moment correlation coefficient in random samples of any size draw fromnon-normal universes". Biometrika38: 219–247. doi:10.1093/biomet/38.1-2.219.[9]Soper, H.E., Young, A.W., Cave, B.M., Lee, A., Pearson, K. (1917). "On the distribution of the correlation coefficient in small samples.Appendix II to the papers of "Student" and R. A. Fisher. A co-operative study", Biometrika, 11, 328-413. doi:10.1093/biomet/11.4.328[10]Wilcox, Rand R. (2005). Introduction to robust estimation and hypothesis testing. Academic Press.[11]Devlin, Susan J; Gnanadesikan, R; Kettenring J.R. (1975). "Robust Estimation and Outlier Detection with Correlation Coefficients" (http:///stable/2335508). Biometrika62 (3): 531–545. doi:10.1093/biomet/62.3.531. .[12]Huber, Peter. J. (2004). Robust Statistics. Wiley.[13]Multivariable Analysis- A Practical Guide for Clinicians. 2nd Edition. (/uk/catalogue/catalogue.asp?isbn=052154985X&ss=exc) Mitchell H. Katz. University of California, San Francisco. ISBN 9780521549851. ISBN 052154985X DOI:10.2277/052154985X[14]/Archive/sci.stat.math/2006-02/msg00171.html[15] A MATLAB Toolbox for computing Weighted Correlation Coefficients (/matlabcentral/fileexchange/20846)[16]Cox, D.R., Hinkley, D.V. (1974) Theoretical Statistics, Chapman & Hall (Appendix 3) ISBN 0412124203Article Sources and Contributors10 Article Sources and ContributorsPearson product-moment correlation coefficient Source: /w/index.php?oldid=406110938 Contributors: AgarwalSumeet, Albmont, Amillar, Arbitrary username,Arcadian, AxelBoldt, Baccyak4H, Beno1000, Bobo192, Bramschoenmakers, Can't sleep, clown will eat me, Chris53516, Cometstyles, Countchoc, Damian Yerrick, Deditos, Delirium, Denfjättrade ankan, DerHexer, Dfarrar, Discospinster, Dpryan, Dr.enh, Drbreznjev, Dysprosia, FrancisTyers, G-J, G716, Garion96, Giftlite, Gringotumadre, Ichbin-dcw, Ignoramibus,Irregulargalaxies, JamesBWatson, JeremyA, Jfitzg, Jmath666, JohnEBredehoft, JorisvS, Jtneill, Juancitomiguelito, Karada, Karl Dickman, Karol Langner, Kinneyboy90, Kku, Kyawtun,Landroni, Ldm, MER-C, Mahahahaneapneap, Male1979, Mc4932, Mcld, Melcombe, Michael Hardy, Mild Bill Hiccup, Mmmready, MrOllie, MrYdobon, NeonMerlin, Notheruser, O18, Parodi, Pearle, Qniemiec, Qwfp, Rajah, Ramkgupta1, RedCoat10, Rgclegg, Rich Farmbrough, Rifleman 82, Rjwilmsi, Skbkekas, Stewartadcock, Sykko, Talgalili, Tdslk, Tellyaddict, Tesi1700, Tomharrison, Trontonian, Versus22, Vinhtantran, Vrenator, WikHead, Wikifriend pt001, Wootini, Xenoglossophobe, Yamiken, Yerpo, 170 anonymous editsImage Sources, Licenses and ContributorsImage:Correlation examples.png Source: /w/index.php?title=File:Correlation_examples.png License: Public Domain Contributors: Original uploader was Imagecreator at en.wikipediaFile:Regression lines.png Source: /w/index.php?title=File:Regression_lines.png License: Creative Commons Attribution-Sharealike 3.0 Contributors: User:Qniemiec Image:correlation significance.svg Source: /w/index.php?title=File:Correlation_significance.svg License: Creative Commons Attribution 3.0 Contributors:User:SkbkekasLicenseCreative Commons Attribution-Share Alike 3.0 Unported/licenses/by-sa/3.0/。

FRM 定量 第二部分0624 [兼容模式]

FRM 定量 第二部分0624 [兼容模式]
a. 9.51 b. 8.60 c. 13.38 d. 7.45
8
Covariance and correlation analysis con't Example: sample covariance calculation
9
3

Covariance and correlation analysis con't
15
5

Ordinary least squares regression
• Ordinary least squares (OLS) estimation is a process that estimates the population parameter Bi with corresponding values for bi that minimize the squared residuals. • Recall the expression ei=Yi-(b0+ b1*Xi), the OLS sample coefficients are those that:
R2 = ρ 2
V ( y ) = β 2V ( x) + V (ε ) Dividingb yV ( y) 1= β 2V ( x) V (ε ) + V ( y) V ( y)
R2 = β 2
V ( x) V ( y)
19
The least squares regression line con't
2
Collecting data
• There are there types of data that are generally available for empirical analysis. • 1. Time series. • 2. Cross-sectional. • 3. Pooled ( a combination of time series and cross-sectional). • Panel data is a special type of pooled data---the same cross-sectional unit is surveyed over time. (also called longitudinal or micropanel data).

Autocorrelation Coefficients and Correlation Immunity of Boolean Functions

Autocorrelation Coefficients and Correlation Immunity of Boolean Functions
e-mails: yutaran@mech.math.msu.su, taran@vertex.inria.msu.ru e-mail: peter-korolev@mtu-net.ru e-mail: stony m@mail.ru
Autocorrelation Coefficients and Correlation Immunity
463
based stream ciphers is correlation immunity introduced by Siegenthaler [13]. Another important properties are nonlinearity, algebraic degree and so on. For Boolean functions used in block ciphers the most important properties are nonlinearity and differential (or autocorrelation) characteristics (propagation degree, avalanche criterion, the absolute indicator and so on) based on the autocorrelation coefficients of Boolean functions. Note that in recent research differential characteristics are considered as important for stream ciphers too. Correlation immunity (or resiliency) is the property important in cryptography not only in stream ciphers. This is an important property if we want that the knowledge of some specified number of input bits does not give a (statistical) information about the output bit. In this respect such functions are considered in [6], [3] and other works. Many works (see for example [5]) demonstrate that correlation immunity and autocorrelation characteristics are in strong contradiction. Some of results in our paper confirm it. Nevertheless, it appears that autocorrelation coefficients of a Boolean function is a power tool for the investigation of correlation immunity and other properties even without a direct relation to differential characteristics. The results of our paper demonstrate it. In Section 2 we give preliminary concepts and notions. In Section 3 we prove −n+3 2n for the absolute indicator of resilient funcnew lower bound ∆f ≥ 2m n+1 tions that improves significantly (for m > (n − 3)/2) the bound of Zheng and Zhang [18] on this value. In Section 4 we prove that the number of nonlinear variables in n-variable (n − k )-resilient Boolean function does not exceed (k − 1)2 k−2 . This result supersedes the previous record n ≤ (k − 1)4k−2 of Tarannikov and Kirienko [16]. As a consequence we give the sufficient condition on m and n that the absolute indicator of n-variable m-resilient function is equal to the maximum possible value 2n . In Section 5 we characterize all possible values of resiliency orders for quadratic functions, i. e. functions with algebraic degree 2 in each variable. In Section 6 we give a complete description of quadratic n-variable m-resilient Boolean functions that achieve the bound m ≤ n 2 − 1. In Section 7 we establish new necessary condition that connects m, n and the weight of an n-variable unbalanced nonconstant mth order correlation immune function and prove that such functions do not exist for m > 0.75n − 1.25. For high orders of m this surprising fact supersedes the well-known Bierbrauer–Friedman bound [8], [1] and was not formulated before even as a conjecture. In Section 8 we prove 1 1 π 8/9 − 1, n ≥ 12, the nonlinearity of an that for m ≥ 1 2 n + 2 log2 n + 2 log2 2 e unbalanced mth order correlation immune function of n variables does not ex1 1 1 π 8/9 1 ceed 2n−1 − 2m+1 , and for m ≥ 2 n+ 3 − 2, 2 log2 n + log2 4 + n + 2 log2 2 e n ≥ 24, this nonlinearity does not exceed 2n−1 − 2m+2 . These facts improve significantly correspondent results of Zheng and Zhang [18] and demonstrate that for higher orders of resiliency the maximum possible nonlinearity for balanced functions is greater than for unbalanced. Along all paper we apply actively autocorrelation and Walsh coefficients for the investigation of correlation immune and resilient Boolean functions. Our new results demonstrate the power of this approach.

Correlation

Correlation

4.逐个计算每个被评价对象指标序列与参考序列对应元素的绝对差值
5. 确定 6.计算关联系数
分别计算每个比较序列与参考序列对应元素的关联系数:(通常ξ取0.5)
7.计算关联度
8.依据各观察对象的关联序,得出综合评价结果.
Example:
1、六位教师评价指标包括:专业素质、外语水平、教学工作量、 科研成果、论文、著作与出勤 2、
Kendall tau:分类变量。如几个评定这对一个项目评定的一致性, 同一个评定这几次等级评定的一致程度。
Spearman:排列后的变量之间的Pearson线性相关系数。
两个变量均为等级变量。不要求二元正态分布。
Partial correlation
In probability theory and statistics, partial correlation measures the degree of association between two random variables, with the effect of a set of controlling random variables removed.
accompanied by an increase in y.
if y always decreases when x increases, the rank correlation coefficients will be −1
Conclusion:
Pearson相关:二元正态分布,连续型变量
It can be shown that the residuals RX coming from the linear regression of X using Z. The same also applies to the residuals RY generating a vector rY. This means that the residuals vector lives on a hyperlane Sz that is perpendicular to z. The desired partial correlation is then the cosine of the angle φ between the projections rX and rY of x and y, respectively, onto the hyperplane perpendicular to z.

实验心理学 脑认知(功能)成像技术与心理学研究——第1部分

实验心理学 脑认知(功能)成像技术与心理学研究——第1部分

Observed first as a 9-yr. old, then later in medical school, that classmates with bulging eyes had great verbal memories.
Later theorized verbal memory was localized in front of brain; more developed memory caused enlarged frontal lobe, which
Brain
The Process of Discovery
Correlation vs. Causation
• Often considered a case study in bad science (socalled pseudoscience)
• But the idea of localization of function is important
Correlation vs. Causation
Correlation: a linear relationship between two variables
positive correlation: as one variable increases, the other variable increases
《Discourse on the Method》
Slide modified from Professor Ernst Pöppel
Brain
The Process of Discovery
Two Complementary Ways of Doing Science:
Exploration (Searching and Finding the Unexpected; Serendipity)

The Redshift-Space Two Point Correlation Function of ELAIS-S1 Galaxies

The Redshift-Space Two Point Correlation Function of ELAIS-S1 Galaxies

a r X i v :a s t r o -p h /0503258v 2 17 M a r 2005Mon.Not.R.Astron.Soc.000,1–??(2002)Printed 2February 2008(MN La T E X style file v2.2)The Redshift-Space Two Point Correlation Function ofELAIS-S1GalaxiesV.D’Elia,1E.Branchini, Franca,2V.Baccetti,2I.Matute,3F.Pozzi,4C.Gruppioni 51INAF-Osservatorio Astronomico di Roma via Frascati 33,Monteporzio-Catone (RM),I-00040Italy2Dipartimentodi Fisica,Universit´a degli Studi “Roma Tre”,Via della Vasca Navale 84,I-00146,Roma,Italy3Max-Planck Institut f¨u r extraterrestrische Physik (MPE),Giessenbachstrasse,Postfach 1312,85741Garching,Germany 4Dipartimento di Astronomia,Universit`a di Bologna,viale Berti Pichat 6,I-40127Bologna,Italy 5INAF-Osservatorio Astronomico di Bologna,via Ranzani 1,I-40127Bologna,ItalyReleased 2005January 14ABSTRACTWe investigate the clustering properties of galaxies in the recently completed ELAIS-S1redshift survey through their spatial two point autocorrelation function.We used a sub-sample of the ELAIS-S1catalog covering approximately 4deg 2and consisting of 148objects selected at 15µm with a flux >0.5mJy and redshift z <0.5.We detected a positive signal in the correlation function that,in the range of separations 1−10h −1Mpc is well approximated by a power law with a slope γ=1.4±0.25and a correlation length s 0=5.4±1.2h −1Mpc,at the 90%significance level.This result is in good agreement with the redshift-space correlation function measured in more local samples of mid infrared selected galaxies like the IRAS PSC z redshift survey.This suggests a lack of significant clustering evolution of infrared selected objects out to z =0.5that is further confirmed by the consistency found between the correlation functions measured in a local (z <0.2)and a distant (0.2<z <0.5)subsample of ELAIS-S1galaxies.We also confirm that optically selected galaxies in the local redshift surveys,especially those of the SDSS sample,are significantly more clustered than infrared objects.Key words:galaxies:clusters:general -galaxies:evolution -cosmology:observation -Large scale structure of Universe -infrared:galaxies1INTRODUCTIONInvestigating the redshift-space distribution of galaxies has long been regarded as a fundamental aspect of observational cosmology.The primary statistical tool for characterizing galaxy clustering is the spatial two point correlation func-tion,ξ(s )since,in the current paradigm of structure for-mation,the galaxy two-point correlation function is directly related to the initial power spectrum of mass density fluc-tuations.This is true as long as galaxies trace the underlying mass density field.However,it is now well established that the clustering of galaxies at low redshift depends on a vari-ety of factors,implying that not all types of galaxies can be regarded as unbiased mass tracers.The clustering of opti-cally selected galaxies has been found to depend on galaxy luminosity (Norberg et al.2002and reference therein),mor-phological and spectral type (Hermit et al.1996,Zehavi et al.2002,Magdwick et al.2003).On the other hand,the clustering of infrared selected galaxies seems to depend ontheir infrared color (Hawkins et al.2001)rather than on luminosity (Szapudi et al.2000).The evidence that different galaxy populations might give different biased pictures of the mass distribution has complicated but also enriched the interpretation of galaxy clustering.Indeed,the very fact that the spatial clustering of galaxies is related to their physical properties represents an important observational test for all theories of galaxy for-mation.In particular,strong constraints on galaxy evolution models can be obtained by measuring the relative clustering of different extragalactic objects as a function of redshifts,for which very deep galaxy samples are required.Several deep redshift surveys of optically selected galax-ies,like the Keck surveys of the GOODS-north (Wirth et al.2004)and GOODS-south (Vanzella et al.2004)fields and the DEEP2redshift survey (Davis et al.2003),are cur-rently being performed.First results based on early data look very promising indeed.The analysis of Coil et al.(2004)has shown that the two-point correlation function of DEEP2galaxies with a median redshift z =1.14is consistent with2V.D’Elia et al.that measured by Adelberger et al.(2003)in a very deep (z∼3)sample of Lyman break galaxies but is significantly smaller than the correlation measured in the local(z∼0) 2dF galaxy sample.Since2dF b j-selected galaxies are known to trace the underlying mass densityfield in a unbiased way (Verde et al.2002,Lahav et al.2002),the smaller correla-tion measured in DEEP2survey implies that this is not a strongly biased sample of objects either,since the clustering of the dark mass is expected to decrease with redshift in a similar way(Coil et al.2004).Mid-infrared selected galaxies also constitutes a very in-teresting population of objects since they are also known to trace the underlying mass distribution in the local universe in an almost unbiased fashion.More precisely,it has been found that thefluctuations in their number density,δg is related to the underlying mass overdensityfieldδm through a simple,linear biasing relationδg∼1.2δm(Tegmark,Zal-darriaga&Hamilton2001and Taylor et al.2002).Another reason why mid-infrared selection is interesting is that lu-minosity in this band is approximately proportional to star-formation rate(independent of dust),thus a mid-infrared sample of galaxies will highlight the distribution of star-formation activity at a particular epoch.Clearly,it would be very important to quantify the clustering evolution of mid-infrared selected objects and compare it with that of optically selected galaxies.This is indeed the main goal of this work in which we analyze the clustering properties of sample of mid-infrared selected galaxies extracted from the ELAIS redshift survey(Oliver et al.2000)and extend-ing out to a redshift z=0.5.According to La Franca et al.(2004),two main spectroscopic classes have been found to dominate the extragalactic population of these objects: star-forming galaxies(from absorbed to extreme starbursts:νLν(15µm)∼108−1011L⊙),which account for75%of the sources,and active galactic nuclei(excluded from this anal-ysis)which account for25%of the sources.About20%of the extragalactic ELAIS sources are dust-enshrouded star-burst galaxies,while passive galaxies are essentially absent from the sample.Our analysis is performed in redshift-space and thus complements the previous work of Gonzalez-Solares et al.. (2004)who measured the angular correlation properties of a similar sample of ELAIS galaxies that from which they have inferred their spatial correlation properties.The outline of this paper is as follows.In section2we describe the ELAIS-S1galaxy redshift survey that we an-alyze in this work.In section3we discuss our method of estimating the two-point correlation function,assess its ro-bustness and evaluate its statistical uncertainties.The main results are presented in section4,and discussed in section 5,in which we also draw our main conclusions.2THE ELAIS S1SAMPLEThe European Large-Area ISO survey(ELAIS,Oliver et al. 2000;Rowan-Robinson et al.2004)is the largest Open Time programme conducted by the ISO satellite(Kessler et al. 1996).It covers an area of12deg2,divided in fourfields (N1,N2and N3in the northern hemisphere and S1in the southern one)distributed across the sky in order to de-crease the biases due to cosmic variance.The survey bands are at6.7,15,90and170µm;the15µm one presents the highest density of galaxies(Serjeant et al.2000,Gruppioni et al.2002,La Franca et al.2004),making it the best choice for a study of their clustering properties.In this work we concentrate on the southern area,S1.This survey is made of nine raster observations,each covering40×40arcmin2.Thefinal analysis catalog at15µm in the S1field has been released by Lari et al.(2001) covering an area of2×2deg2centered atα(2000)=00h 34m44.4s,δ(2000)=−43◦28′12′′.It includes462mid-IR sources down to aflux limit of0.5mJy.We have restricted our analysis to a highly reliable subsample of406objects (La Franca et al.2004).A detailed description of the optical classification of the ELAIS-S1sources,size and completeness function of the ar-eas used in our study,as well as the observed counts for each class of sources(normal galaxies,Starburst galaxies and AGN)are presented by La Franca et al.(2004).The mea-sure of the evolution of star-forming galaxies has been inves-tigated by Pozzi et al.(2004)and Gruppioni et al.(2005), while afirst estimate of the luminosity function for type-1 AGN has been presented by Matute et al.(2002)1.The central raster,S1R5)which covers0.55deg2and reaches a20%completeness atfluxes of 0.7mJy,and b)the remaining area(S1R5raster and the remaining100galaxies to the surrounding areas S1R5raster.The continuous line histograms in both panels of Fig. 2shows the redshift distribution of the ELAIS-S1galaxies in the S1Rest(bottom)rasters.Gonzalez-Solares et al..(2004)have shown that the redshift distri-bution of ELAIS objects obtained from follow-up spectro-scopic observations and photometric redshifts is consistent with that predicted from the ELAIS luminosity function of Pozzi et al.(2004).This redshift distribution will used in the next section to estimate the galaxy correlation function and to construct the mock ELAIS catalogs.1Data and related papers about the ELAIS southern survey are available at:http://www.fis.uniroma3.it/∼ELAISMonthly Notices:L a T E X2εguide for authors3Figure 1.The angular distribution of ELAIS-S1sources.Thesolid square is the central raster S1R5raster(top panel)and in S1n RN RR(s) n R N RR(s)4V.D’Elia et al. w i=1H o z0dz′Ωm(1+z′)3+ΩΛ,(3)where r is the comoving distance,and then added the line of sight component of the particle peculiar velocity.Finally, a population of mock galaxies have been extracted from the particles through a Montecarlo rejection procedure designed to match the observed redshift distribution of real ELAIS galaxies in both the inner40×40arcmin2S1Rest-like samples.This procedure has been re-peated to obtain30mock ELAIS-S1samples,containing Figure3.Averageξ(s)measured in the30mock ELAIS-S1cata-logs(filled dots)and the rms scatter around the mean(errorbars). Open dots shows the“true”ξ(s)of all particles in the simulation contained in a region with the same volume of the ELAIS-S1 sample.The straight line represents the best power lawfit to the “true”ξ(s)in the range1−10h−1Mpc.The values of the best fit parameters are also shown.∼120objects each,for which we have evaluated the two-point correlation function.The mean redshift distribution of fake objects in the30catalogs is shown in of Fig.2(dashed line histogram)for the S1Rest(bot-tom)rastersThefilled dots in Fig.3shows the averageξ(s)in the 30mock catalogs.The errorbars represent the rms scatter around the mean values.They constitutes our estimate of er-rors in the measurement of two point correlation function of the real ELAIS-S1galaxies These errors account for cosmic variance and sample noise,that constitutes the main source of uncertainty.The large size of the sample guarantees that the’integral constraint’correction is only5%(Gonzalez-Solares et al.2004)and thus can be neglected in the error budget.The open dots in Fig.3shows the“true”ξ(s)of the simulation,measured by considering all particles in one of our ELAIS-like samples.The“true”ξ(s)is consistent with the averageξ(s)of the mocks,indicating that our method for estimatingξ(s)is indeed unbiased.The mock ELAIS-S1catalogs mimic the geometry and selection effects of the real sample and account for the spa-tial clustering and its evolution in aflatΛCDM universe. However,they are not guaranteed to reproduce the cluster-ing properties of the ISO galaxies unless,of course,these trace the underlying mass densityfield out to z=0.5.As we have verified a posteriori this is indeed the case, in the sense that the“true”ξ(s)in the range1−10h−1Mpc is well approximated by a power law whose bestfit param-eters,displayed in Fig.3are consistent within the errors with those determined in the analysis of the real ELAIS-S1 sample presented in the following section.Monthly Notices:La T E X 2εguide for authors5Figure 4.The redshift-space correlation function for the ELAIS-S1galaxy sample (filled dots).Error bars have been calculatedusing the 30ELAIS-like mock catalogs.The solid line represents the power law best fit ξ(s )=(s/s 0)−γto the data in the range 1−10h −1Mpc.The best fit parameters are indicated in the plot.4RESULTSThe filled dots plotted in Fig.4represent the two-point cor-relation function of the ELAIS-S1galaxies computed us-ing the LS estimator.The errorbars,evaluated from the 30ELAIS-S1mock catalogs,are the same shown in Fig.3.Clearly,ξ(s )is well approximated by a power law out to separations of 10h −1Mpc.In the range 1−10h −1Mpc the best fit power law model ξ(s )=(s/s 0)−γ,has a correlation length of s 0=5.4±1.2h −1Mpc and a slope γ=1.45±0.25,both determined at the 90%confidence level.Including the correlation of galaxy pairs at separation s =0.6h −1Mpc,also shown in the figure,does not modify this result appre-ciably.Breaking down the sample by redshift does not change results significantly either.Indeed,in the local sample com-posed by 82objects at z <0.2ξ(s )is still well approxi-mated by a power law in the range 1−10h −1Mpc with best fit parameters s 0=5.4±1.6h −1Mpc and γ=1.6±0.4.These values agree with those found for the distant sample of 66objects at at 0.2 z <0.5for which we have found s 0=5.1±1.6h −1Mpc and γ=1.4±0.4,Finally,we have verified that evaluating errors from 100bootstrap realizations of the ELAIS-S1sample rather than from the mock catalogss does not change significantly the results as the correlation length only decreases by ∼10%and the slope becomes ∼4%flatter.5DISCUSSION AND CONCLUSIONSIn this work we have evaluated the redshift-space two point correlation function of mid-infrared selected galaxies in the deep (z 0.5)ELAIS-S1catalog.We found a significant,positive correlation signal at separations 10h −1Mpc,where the two point correlation function is well approx-imated by a power law model ξ(s )=(s/s 0)−γwith a correlation length of s 0=5.4±1.2h −1Mpc and a slope γ=1.45±0.25,with errorbars referring to a 90%confi-dence level.These results have been obtained using the LS estimator for ξ(s )and by evaluating the errors from a set of 30mock ELAIS-S1catalogs.These results are robust,in the sense that they do not change significantly when varying the method of estimating ξ(s )(see section 3.1)or when using different strategies to assess the errors (sections 3.2and 4).It is interesting to compare our results with those obtained from similar analysis of galaxy clustering in redshift-space.Table 1shows our result together with the corresponding results obtained from some of the ma-jor galaxy redshift surveys,characterized by their pass-band/wavelength of selection (column 2)and the redshift ranges they cover (column 3).It is worth stressing that de-viations of the two point correlation function from a pure power law shape are more serious in redshift-space than in real-space,making it difficult to compare results obtained from the analyses of different galaxy samples.To ensure a fair comparison,all best fit parameters listed in columns 4and 5of Table 1refer to ranges of separations where all measured ξ(s )are well approximated by a power law.These ranges turned out to be very close to that of [1−10]h −1Mpc considered in our analysis,although some of the parameters in the Table have been obtained by pushing the estimate of ξ(s )down to scale as small as 0.1h −1Mpc (the 2dFGRS case)or up to separations as large as 16.4h −1Mpc (as in the case of the LCRS sample).Our results are fully consistent with those obtained from the analysis of the PSC z survey (Hawkins et al.2001),that consists of ∼15,000IRAS galaxies selected at 60µm ,i.e.in a mid-infrared band similar to that of ELAIS galaxies.This sample is,however,much more local than ours and thus the agreement between the two results indicates that the clustering properties of mid-infrared selected objects do not evolve significantly between z =0and z =0.3.This conclu-sion is corroborated by the consistency found between the two measurements of ξ(s )performed in the local (z <0.2)and distant (0.2 z <0.5)ELAIS-S1subsamples.Unfortu-nately,the large uncertainties in our estimate of ξ(s ),that mainly result from the sparseness of the ELAIS-S1galaxy catalog,do not allow to set strong constraints on the clus-tering evolution.The lack of significant evolution in the ELAIS galaxy clustering has already been noticed by Gonzalez-Solares et al.(2004).In their analysis,that consisted in deprojecting the angular correlation function of ELAIS galaxies via Lim-ber equation,they have measured a real-space two point cor-relation function with a slope (γ=2.04±0.18)and a corre-lation length (r 0=4.3+0.4−0.7h−1Mpc)that are fully consistent with those measured in the PSC z catalog (γ=2.04±0.18and r 0=∼3.7h −1Mpc,Jing,B¨o rner &Suto (2002)).It is worth stressing that difference between our result and that of Gonzalez-Solares et al.(2004)originates from systematic redshift space distortions that affect our analysis and result in a shallower slope and a larger correlation length of the two point-correlation function.Focusing on the mid-infrared objects is of considerable interest since they trace the underlying mass distribution in the local universe.The consistency that we have found6V.D’Elia et al.Table1.Clustering of Different Galaxy Redshift Surveys:the Parameters of the Power Law ModelELAIS-S1115µm0.0-0.55.40±1.201.45±0.25PSC z260µm0.004-0.14.77±0.201.30±0.04CfA23B-band0.0-0.05∼7.5∼1.6ORS4B-band0.0-0.0277.60±1.201.60±0.10LCRS5R-band0.033-0.156.3±0.31.86±0.03SDSS5r⋆-band0.019-0.13∼8.0∼1.22dFGRS6b j-band0.01-0.206.82±0.281.57±0.071This analysis2Hawkins et al.(2001)3de Lapparent,Geller&Huchra(1988)4Hermit et al.(1996)5Tucker et al.(1997)6Zehavi et al.(2002)7Hawkins et al.(2003)Monthly Notices:L a T E X2εguide for authors7 Verde,L.et al.2002,MNRAS,335,432Wirth,G.D.et al.2004,AJ,127,3121Zehavi,I.et al.2002,ApJ,571,172。

计量经济学中英文词汇对照

计量经济学中英文词汇对照

Controlled experiments Conventional depth Convolution Corrected factor Corrected mean Correction coefficient Correctness Correlation coefficient Correlation index Correspondence Counting Counts Covaห้องสมุดไป่ตู้iance Covariant Cox Regression Criteria for fitting Criteria of least squares Critical ratio Critical region Critical value
Asymmetric distribution Asymptotic bias Asymptotic efficiency Asymptotic variance Attributable risk Attribute data Attribution Autocorrelation Autocorrelation of residuals Average Average confidence interval length Average growth rate BBB Bar chart Bar graph Base period Bayes' theorem Bell-shaped curve Bernoulli distribution Best-trim estimator Bias Binary logistic regression Binomial distribution Bisquare Bivariate Correlate Bivariate normal distribution Bivariate normal population Biweight interval Biweight M-estimator Block BMDP(Biomedical computer programs) Boxplots Breakdown bound CCC Canonical correlation Caption Case-control study Categorical variable Catenary Cauchy distribution Cause-and-effect relationship Cell Censoring

QM1 – Week 2Correlation

QM1 – Week 2Correlation
2
4. Scatterplots and Correlation
• The higher the real wage in 1935 the lower the industrial production in 1935 • There is a negative correlation between real wage and industrial production
QM1 – Week 2 Correlation
Dr Alexander Moradi University of Oxford, Dept. of Economics & GPRG/CSAE Email: alexander.moradi@
1
4. Correlation
• Correlation describes the relationship between two or more variables • Correlation IS NOT causation!!! • Example: Income and health (life expectancy, morbidity) are correlated
• pwcorr varlist
• spearman varlist
Use STATA’s help function (Help/Stata command…) to learn more about the options
12
4. Homework Exercises
1. Read Chapter 3 of Feinstein & Thomas
8
4. Spearman’s Rank Correlation Coefficient
  1. 1、下载文档前请自行甄别文档内容的完整性,平台不提供额外的编辑、内容补充、找答案等附加服务。
  2. 2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
  3. 3、如文档侵犯您的权益,请联系客服反馈,我们会尽快为您处理(人工客服工作时间:9:00-18:30)。

2. Theory. In order to investigate monopole currents one has to project SU(3) onto its abelian degrees of freedom, such that an abelian U(1) U(1) theory remains 5]. This aim can be achieved by various gauge xing procedures. We employ the so-called maximal abelian gauge which is most favorable for our purposes. After gauge xing pure QCD may be regarded as a theory of color charges and color magnetic monopoles. A gauge transformation of a gauge eld element U(x; ) is given by ~ U(x; ) = g(x)U(x; )gy (x + ^) ; (1) where g(x) 2 SU(3). The maximal abelian gauge is imposed by maximizing the functional X ~ (2) R = jUii (x; )j2:
We analyze the relation between instantons and abelian projected monopoles in both phases of pure QCD by calculating local correlation functions between topological charge densities and monopole densities. On an 83 4 lattice, it turns out that topological quantities are correlated approximately two lattice spacings. The monopole-instanton correlations are rather insensitive under cooling of gauge elds.
hep-lat/9510024 13 Oct 1995
1. Introduction. It is assumed that gauge eld con gurations with non-trivial topology can explain the low-energy properties of QCD and thus could provide a solution to the con nement problem. There are two di erent kinds of topological objects which seem to be important for the connement mechanism: color magnetic monopoles and instantons. Color magnetic monopoles play the essential role in the dual superconductor hypothesis 1]. In this picture con nement emerges by condensation of abelian monopoles via the dual Meissner e ect. There is strong evidence from lattice calculations that the idea of dual superconductivity is an adequate description of the con nement phenomenon. On the other hand the role of instantons with respect to con nement is not so clear. It is suggested that instantons can only cause con nement in QCD by forming a socalled instanton liquid 2]. A fascinating question is whether color magnetic monopoles and instantons are related or not. In a recent study we analyzed the distribution of the topological charge density and the monopole density around static color sources 3]. We observed that both topological quantities show qualitatively the same behavior, namely a suppression in the vicinity of the external sources. Motivated by this similarity we calculate local correlation functions between the monopole density and the topological charge density 4]. Results for the size of the correlations between topological objects are reported.
23@f (x+ ^;
)
x; ;i
i
mod 2
Supported in part by "Fonds zur Forderung der wissenschaftlichen Forschung" under Contract P9428-PHY.
2 where ui (2) denotes a product of abelian links ui (x; ) around a plaquette 2 and f(x + ^; ) is an elementary cube perpendicular to the direction with origin x + ^. The magnetic currents form closed loops on the dual lattice as a consequence of monopole current conservation. Finally the local monopole density is given by X (5) (x) = 3 1 4V4 ;i jmi (x; )j : For the implementation of the topological charge on the lattice there exists no unique discretization. In this work we restrict ourselves to the so-called eld theoretic de nitions which approximate the topological charge in the continuum 6] g2 q(x) = 32 2 Tr F (x)F (x) ; (6) in the following ways: 4 X 1 (P;H ) ~ Tr O(P;H ) ; (7) q (x) = 24 32 2 ;:::= 1 with O(P ) = U (x)U (x) ; (8) for the plaquette prescription and O(H ) = U(x; )U(x+^; )U(x+^+^; ) U(x+^ +^ +^; )U y(x+^ +^+^ ; ) U y(x+^+^ ; )U y(x+^ ; )U y(x; ); (9) for the hypercube prescription. The lattice and continuum versions of the theory represent di erent renormalized quantum eld theories, which di er from one another by nite, non-negligible renormalization factors 7]. A simple procedure that enables one to get rid of renormalization constants, while preserving physical information contained in lattice con gurations, is the cooling method. In our investigation we have employed the so-called \Cabbibo{Marinari method". To measure the correlations between topological quantities we calculate the functions hq(0)q(d)i ; hq2 (0)q2 (d)i ; h (0) (d)i ; h (0)q2 (d)i : (10) Since topological objects with opposite sign are equally distributed, we correlate the monopole density with the square of the topological charge density. 3. Results. Our simulations were performed on an 83 4 lattice with periodic boundary conditions using the Metropolis algorithm. The observables were studied in pure QCD with the plaquette action both in the con nement and decon nement phase at inverse gluon coupling = 6=g2 = 5:6 and 5:8, respectively. We made 50000 iterations and measurements were taken after every 50th iteration. Each of these 1000 congurations was rst cooled and then subjected to 300 gauge xing steps enforcing the maximal abelian gauge. The correlation functions between topological quantities according to Eq. (10) are shown in Fig. 1 for several cooling steps at = 5:6. They are normalized after the subtraction of their cluster values. The hypercube de nition is used for q. The range of the instanton auto-correlations qq and q2 q2 which are originally -peaked grows rapidly with cooling re ecting the occurance of extended instantons 8]. In contrast the correlation decreases since monopole loops become dilute with cooling. The q2 -correlation seems rather insensitive to cooling and clearly extends over more than two lattice spacings, indicating some non-trivial local correlation between monopoles and topological charges. Fig. 2 presents row q2 -correlations for the plaquette and the hypercube de nition of the topological charge density in the con nement (l.h.s.) and the decon nement phase (r.h.s.) for 6 and 11 cooling steps. The agreement between the two de nitions is almost perfect. The absolute value of the correlations is smaller in the decon nement phase since both instantons and monopoles become dilute. But there are practically no differences between the q2 -correlations in the two phases after normalization (not shown). This holds also for the other topological correlation functions. In Fig. 3 the sumP the squared topological of charge density K := x q2 (x) (hypercube de nition) is plotted against the number of monopoles P Nmon = x (x) for each con guratinctions between monopoles and instantons
相关文档
最新文档