ESTIMATES OF VARIANCE COMPONENTS IN RANDOM EFFECTS META-ANALYSIS: SENSITIVITY TO VIOLATIONS OF NORMALITY AND VARIANCE HOMOGENEITY

Size: px
Start display at page:

Download "ESTIMATES OF VARIANCE COMPONENTS IN RANDOM EFFECTS META-ANALYSIS: SENSITIVITY TO VIOLATIONS OF NORMALITY AND VARIANCE HOMOGENEITY"

Transcription

1 ESTIMATES OF VARIANCE COMPONENTS IN RANDOM EFFECTS META-ANALYSIS: SENSITIVITY TO VIOLATIONS OF NORMALITY AND VARIANCE HOMOGENEITY Jeffrey D. Kromrey and Krstne Y. Hogarty Department of Educatonal Measurement and Research, Unversty of South Florda, Tampa, FL 3360 KEY WORDS: Meta-analyss, Varance Components, Random Effects Models Meta-analyss s a statstcal technque through whch nformaton obtaned from a collecton of ndependent studes s analyzed and syntheszed, potentally provdng a more complete representaton of the phenomenon under nvestgaton. Models for meta-analyss may be roughly dvded nto those based upon fxed effects and those based upon random effects (Feld, 00; Hedges, 994; Hedges & Vevea, 998; Raudenbush, 994). The prmary goal of meta-analyss s to obtan estmates of populaton effect szes and confdence bands around those estmates. For example, recent research n the area of educatonal reform related to the Natonal Scence Foundaton s Systemc Intatves (Kromrey et al., 00) has nvestgated the use of effect szes as ndces of the effectveness of systemc change. These efforts have been drected towards syntheszng student outcome dfferences across subject areas, grade levels, and years of partcpaton. Because sample effect szes obtaned for a metaanalyss typcally present dfferent magntudes of estmaton error, weghted means and varances are used to obtan the estmates of populaton effect szes and confdence bands. For fxed effects models, these weghts are gven by v = σ where σ = the estmaton varance of the th effect sze. In contrast, for random effects models, the weghts used are gven by w = σ τ + where τ = σ = the populaton varance n effect szes. Because τ s not known, t must be estmated from the observed sample effect szes. Three estmators of τ have been suggested n the lterature. One estmator s derved from the observed varance of the sample effect szes and the average of the estmaton varances. s σ! σ = k where s = the ordnary sample varance of effect szes, and k = the number of studes n the metaanalyss. A second estmator, and the most frequently employed n meta-analyss, s derved from the obtaned value of Q (the frequently used test of homogenety of effect szes, Hedges & Olkn, 985). ( k ) Q = c v where c = v. v Fredman (000) presented the analytcal dervaton of the varance of these estmators under the assumpton of normalty and demonstrated that although both estmators are unbased, the varance of s less than that of σ! when τ s close to zero, and hence s a more effcent estmator. Conversely, when τ s large, σ! s a more effcent estmator. A maxmum lkelhood approach to the estmaton of τ was presented by Bggerstaff and Tweede (997) based on the lkelhood equaton: k k k ( y ) ;, = exp = σ + τ = σ + τ ( ) ( ) l y τ π where the y are the observed sample effect szes and = the populaton mean effect sze. The estmaton of τ by maxmum lkelhood (.e., τ ) also provdes an estmate of the varance of τ whch allows the constructon of confdence bands around τ. Further, Bggerstaff and Tweede (997) derved the varance of, and provded a method for constructng confdence bands around ths frequently used estmator. Although the samplng errors of and τ have been derved, an assumpton of normalty was requred. The assumpton of normalty for many educatonal varables s often a tenuous one (Mccer, 989). Further, prevous research (Hogarty & Kromrey, 00) has suggested that commonly used 963

2 effect sze estmates themselves (e.g., standardzed mean dfferences) become based under condtons of non-normalty or heterogeneous varances n the prmary studes. PURPOSE OF THE STUDY The behavor of the three estmators of τ (the σ!,, and τ ; and the nterval pont estmates estmates constructed usng andτ ) under condtons of non-normalty and varance heterogenety cannot be determned analytcally. For such problems, Monte Carlo methods must be appled for the nvestgaton of ther samplng behavors. The current study provdes estmates of the bas and samplng error of these statstcs under condtons n whch the assumptons of normalty and homogenety of varance have been volated. METHOD Ths research was a Monte Carlo study desgned to smulate meta-analyses. The use of smulaton methods allows the control and manpulaton of research desgn factors and the ncorporaton of samplng error nto the analyses. Observatons n prmary studes were generated under known populaton condtons, then prmary studes were combned to smulate meta-analyses. Each prmary study conssted of two groups of observatons (e.g., a treatment and control group). For each prmary study, the Hedges g effect sze was calculated based on the smulated data (Hedges & Olkn, 985). The values of Hedges g obtaned from the samples n each meta-analyss were then used to estmate τ va the three methods descrbed above (.e., σ!,, and τ ). The Monte Carlo study ncluded sx factors n the desgn. These factors were (a) populaton value of τ ( 0,.0,.33,.50, and.00), (b) the number of prmary studes n each meta- analyss (5, 0, 50, and 00), (c) the sample szes of the two groups n each prmary study (wth sample szes rangng from 5 n each group to 00 n each group, as well as unbalanced condtons), (d) group varances n the prmary studes (varance ratos of :, :4, and :8, as well as a homogeneous varance condton), (e) populaton dstrbuton shape n the prmary studes (condtons wth skewness and kurtoss values, respectvely, of 0,0 (.e., normal dstrbuton);,3;.5,5;,6; and 0,5), and (f) the magntude of the populaton mean effect sze ( = 0,.0,.50,.80). Ths research was conducted usng SAS/I verson 8.. Condtons for the study were run under Wndows 98. Normally dstrbuted random varables were generated usng the RANNOR random number generator n SAS. A dfferent seed value for the random number generator was used n each executon of the program. For condtons nvolvng non-normal populaton dstrbutons, the non-normal data were produced by transformng the normal random varates obtaned from RANNOR usng the technque descrbed by Fleshman (978). The program code was verfed by hand-checkng results from benchmark datasets. For each condton examned n the Monte Carlo study, 5000 meta-analyses were smulated. The use of 5000 replcatons provdes a maxmum 95% confdence nterval wdth around an observed proporton that s ± 0.04 (Robey & Barckowsk, 99). The relatve effectveness of the estmators was evaluated n terms of the emprcally estmated statstcal bas and standard errors for the pont estmators, and n terms of confdence band coverage and mean confdence band wdth for the nterval estmates. RESULTS Statstcal Bas and Standard Error The dstrbutons of the statstcal bas estmates for the three pont varance estmators across all condtons examned n ths study are dsplayed as box and whsker plots n Fgure. Examnaton of ths fgure revealed a host of condtons n whch all three methods evdenced postve bas, although the observed varance ndcator appeared to exhbt both greater average bas and somewhat more varablty n bas across the condtons examned. Bas n Estmate Observed Varance Q Based Maxmum Lkelhood Varance Estmator Fgure. Dstrbutons of Statstcal Bas Across Condtons Examned n the Monte Carlo Study Fgure llustrates the dstrbuton of the standard errors of the pont estmators across the condtons examned. The standard errors for all three estmates were reasonably smlar wth each 964

3 estmator evdencng errors of consderable magntude for a number of the condtons examned. These errors were somewhat more pronounced for the observed varance estmator. Evdence of statstcal bas and the assocated samplng error, were further nvestgated n relaton to the central desgn factors n the study. Standard Error Observed Varance Q Based Maxmum Lkelhood Varance Estmator Fgure. Dstrbutons of Standard Error Across Condtons Examned n the Monte Carlo Study Performance wth normal dstrbutons and homogeneous varances. The estmates of bas and standard errors across condtons n whch these two crtcal assumptons are met suggested that the observed varance estmator σ! evdenced the least statstcal bas, but the largest standard errors. Across ths set of condtons, the bas of σ! dd not exceed.07. Further, bas was greatest under condtons wth small sample szes n the prmary studes and large values of τ. Wth large samples, the bas n ths estmator was essentally zero, regardless of the number of studes ncluded n the meta-analyss. In contrast, wth τ > 0, the estmators and τ evdenced substantal bas under condtons of small k or small n. However, the standard errors of these estmators were notably smaller than that of the observed varance estmator. In general, the samplng error of τ was slghtly smaller than that of. The bas of the maxmum lkelhood estmator reached as large as -.38, wth τ =, under small samples, whle that of reached -.3 n these condtons. Wth large k and large n, the maxmum lkelhood estmator evdenced very lttle bas, but the estmator remaned substantally based wth large values of τ. For example, wth τ =, k = 00 and n = n = 00, the bas of was 0.6. Performance wth normal dstrbutons and heterogeneous varances. Under condtons of normal dstrbutons and heterogeneous varances n the prmary studes, the unbased nature of the observed varance estmator was no longer evdent as substantal bas was seen n these estmates, especally under condtons wth small samples n the prmary studes (wth bas reachng as large as.06, wth k = 5, n = 6, 4 and a varance rato of :8). Wth large values of n, however, ths approach provded reasonably unbased estmates f sample szes were equal. In contrast, the and τ estmators evdenced substantal bas n these condtons as the magntude of τ ncreased, although the bas of was typcally smaller than that of τ. A notable excepton occurred wth large values of k and large, equal values of n. In these condtons, the bas of τ dd not exceed.0 n absolute value, whle the bas of reached as large as -.6 wth τ =. Performance wth non-normal dstrbutons and homogeneous varances. For the condtons smulated wth non-normal dstrbutons and homogeneous varances n the prmary studes, the performance of the observed varance estmator depended upon the sample szes of the prmary studes (wth small samples, substantal postve bas was evdent as τ ncreased; wth large samples, relatvely unbased results were obtaned). For condtons smulated wth skewness = and kurtoss = 6, for example, the estmated bas reached as hgh as.69 wth τ = and n = n = 5, but dd not exceed.03 wth large samples. For small samples n the prmary studes, both and τ evdenced superor performance, wth the latter showng slghtly less bas than the former n most condtons. However, wth large n and small k, τ showed notably more bas than, whle wth both k and n large, τ evdenced less bas. For example, wth τ =, k = 5 and n = n = 00, the bas of τ was -.9, whle that of was only In contrast, for the same condton but k = 00, the bas of τ was.0, whle that of was -.4. Performance wth non-normal dstrbutons and heterogeneous varances. The results for heterogeneous varance and non-normal dstrbutons suggested that under condtons of small n, the observed varance estmator performed very poorly regardless of the number of studes n the metaanalyss, wth the estmated bas exceedng.00 for 965

4 the most extreme condtons of varance heterogenety. However, the bases of the other estmators were also substantal under these condtons, especally when heterogeneous varances were pared wth unequal sample szes. As expected, the bas ncreased as the degree of varance heterogenety ncreased. Wth large samples szes, however, the observed varance estmator evdenced the least bas wth equal sample szes or wth postve parngs of sample sze wth populaton varance (wth the bas not exceedng.09 n these condtons). In contrast, wth negatve parngs of sample sze and varance, the bas of ths estmator reached as hgh as.36 n the most extreme condtons of heterogenety. For these condtons, the maxmum lkelhood method performed best wth small values of k (wth bas not exceedng.0), whle performed best wth large k (wth bas not exceedng.08). Confdence Band Coverage and Band Wdth After conductng an analyss of the relatonshps among the pont estmators and the central research desgn factors, we turned our attenton to an examnaton of these relatonshps wth regard to the two confdence nterval estmaton methods. Ths analyss ncluded an examnaton of both the estmated confdence band coverage and the estmated confdence band wdth. The dstrbutons of 95% confdence band coverage across the condtons examned n ths study are dsplayed as box and whsker plots n Fgure 3. As s evdent n ths fgure, both varance estmators provded poor band coverage for a vast number of the condtons examned, wth the maxmum lkelhood estmaton method evdencng both poorer average coverage and consderably more varablty n coverage probablty. Confdence Band Coverage Probablty Q Based Maxmum Lkelhood Varance Estmator Fgure 3. Dstrbutons of Confdence Band Coverage Probablty Across Condtons Examned n the Monte Carlo Study The dstrbutons of confdence band wdths across the condtons examned n ths study are presented n Fgure 4. As s evdent n ths fgure, the Q-based method dsplayed more varablty n the wdths of the confdence ntervals. Mean Confdence Band Wdth Q Based Maxmum Lkelhood Varance Estmator Fgure 4. Dstrbutons of Confdence Band Wdths Across Condtons Examned n the Monte Carlo Study As wth the examnaton of statstcal bas and standard errors of the pont estmates, the confdence band coverage probabltes and confdence band wdths were further nvestgated n relaton to the central desgn factors n the study. Performance wth normal dstrbutons and homogeneous varances. The estmates of confdence band coverage and wdth under selected condtons examned wth normal dstrbutons and homogeneous varances suggested that the confdence bands constructed usng showed excellent coverage of the parameter τ across prmary study sample szes and true values of the parameter for condtons wth small k. However, the confdence bands that provded ths coverage were exceptonally large (reachng as wde as.0 wth the smallest samples, n = 5). Wth larger values of k, the confdence band wdth decreased substantally, but the qualty of the coverage probablty deterorated as well, droppng as low as.6 wth n = 5 and τ =. The confdence ntervals constructed usng τ showed poor performance wth small k and wth small n. Wth large values of k and n the coverage probabltes exceeded.90, but dd not reach the nomnal.95 wth τ > 0. Performance wth normal dstrbutons and heterogeneous varances. The confdence band coverage and wdth for normal dstrbutons wth heterogeneous varances suggested smlar patterns. 966

5 As wth the homogeneous varance condtons, provded confdence bands wth excellent coverage under small k condtons, but the bands were panfully large. In contrast, τ evdenced very poor coverage wth small k and small n. Only under condtons wth both large k and large, balanced n dd the bands computed usng τ show a reasonable degree of coverage (rangng from.9 to.99 wth k = 00 and n = n = 00). Wth unbalanced samples, however, the coverage probabltes of the confdence bands computed usng τ deterorated, reachng as low as.36 under the most extreme condtons of varance heterogenety. Performance wth non-normal dstrbutons and homogeneous varances. The estmates of confdence band coverage and wdth under condtons examned wth non-normal dstrbutons and homogeneous varances suggested that provded good confdence band coverage wth small k, and wth large k coupled wth small n. However, wth large k and large n, the confdence band coverage deterorated. In contrast, τ provded relatvely poor confdence band coverage, except under the large k and large n condtons (where coverage probabltes ranged from.94 to.97). Performance wth non-normal dstrbutons and heterogeneous varances. Fnally, the results from non-normal populaton dstrbutons combned wth heterogeneous varances ndcated that the confdence band coverage of both methods was exceptonally poor n many crcumstances. For example, among the small k condtons, the confdence bands computed usng evdenced coverage probabltes as low as.70 (wth n = 0, 80, τ = 0, and a varance rato of :8). Smlarly, the coverage probabltes of confdence bands constructed usng τ reached as low as.65 (wth n = 80, 0, τ = 0, and a varance rato of :8). Wth condtons characterzed by large k and small n, the confdence band coverage of both methods was notably worse (becomng as low as.0 for confdence bands, and as low as.4 for τ confdence bands). Even wth the large k and large n condtons, confdence band coverage was very poor n many condtons. DISCUSSION Accordng to Freedman (000), under the assumpton of normalty, the Q-based estmator ( ) s a more effcent estmator when τ s close to 0, whereas the observed varance estmator ( σ! ) should provde more effcent estmates wth large values of the populaton value of τ. Clearly, ths pattern was not observed n our analyss of the behavor of the pont estmates under varance heterogenety and non-normalty. In general, the observed varance estmator evdenced notably poorer performance across all condtons examned. In contrast, the Q-based estmator and the maxmum lkelhood estmator provded smlar performance n terms of statstcal bas and standard errors, but showed notable dfference n confdence band coverage probabltes and confdence band wdths. Although both technques provded poorer band coverage under varance heterogenety n the prmary studes, the average performance of the Q- based estmator was consstently more accurate than that of the maxmum lkelhood estmator. The Q- based estmator also provded better band coverage under condtons wth small numbers of studes n the meta-analyss. However, ths superor confdence band coverage comes at a cost of wder confdence bands. The bands obtaned usng the Q-based estmator were consstently wder than those obtaned wth the maxmum lkelhood estmator, often beng several tmes the wdth. All of the estmators examned n ths study evdenced extreme senstvty to volatons of the assumptons of normalty and varance homogenety. Ths senstvty underscores the mportance of assessng the tenablty of these assumptons n prmary studes. However, the estmaton of τ s usually not the prmary goal of random-effects metaanalyss rather, ths statstc s used to obtan the weghts employed n the calculatons of means and varances of effect szes. The mpact of the bas n estmatng τ on these nferences (e.g., the estmaton of populaton mean effect szes) requres further nvestgaton. Meta-analyss has become ncreasngly mportant for the synthess of research results n a varety of felds, ncludng educaton, the behavoral scences and medcne. The accuracy of nferences derved from meta-analyss depends upon the approprate applcaton of statstcal tools. As the use of meta-analytc methods becomes more commonplace, researchers must reman mndful of the lmtatons of certan estmates. The results of ths study underscore the need to exercse cauton n the nterpretaton of the results obtaned from random effects models. The estmates of τ appear to be substantally affected by volatons of the assumptons of normalty and varance homogenety n prmary studes. Ths research furnshes valuable nformaton about the senstvty of estmates of the 967

6 random effect varance and provdes gudance regardng the choce among alternatve methods. Most mportantly, these results hghlght the need for the development of meta-analytc methods that are robust to volatons of these assumptons. ACKNOWLEDGEMENTS Ths work was supported, n part, by the Unversty of South Florda and the Natonal Scence Foundaton, under Grant No. REC The opnons expressed are those of the authors and do not reflect the vews of the Natonal Scence Foundaton or the Unversty of South Florda. REFERENCES Bggerstaff, B. J. & Tweede, R. L. (997). Incorporatng varablty n estmates of heterogenety n the random effects model n meta-analyss. Statstcs n Medcne, 6, Feld, A. P. (00). Meta-analyss of correlaton coeffcents: A Monte Carlo comparson of fxed- and random-effects methods. Psychologcal Methods, 6, Fleshman, A. I. (978). A method for smulatng non-normal dstrbutons. Psychometrka, 43, Fredman, L. (000). Estmators of random effects varance components n meta-analyss. Journal of Educatonal and Behavoral Statstcs, 5, -. Hedges, L. V. (994). Fxed effects models. In H. Cooper & L. V. Hedges (Eds.), The handbook of research synthess (pp ). New York: Russell Sage Foundaton. Hedges, L. V. & Olkn, I. (985). Statstcal methods for meta-analyss. Orlando, FL: Academc Press. Hedges, L. V. & Vevea, J. L. (998). Fxed- and random-effects models n meta-analyss. Psychologcal Methods, 3, Hogarty, K. Y. & Kromrey, J. D. (00, Aprl). We ve been reportng some effect szes: Can you guess what they mean? Paper presented at the annual meetng of the Amercan Educatonal Research Assocaton, Seattle. Kromrey, J. D., Ferron, J. M, Parshall, C. G., Hogarty, K. Y., Grnnell, L., Hess, M. R., Lee, R., Romano, J., Sentovch, C., Watson, F., Dawkns, G. & Nles, J. (00, Aprl). Evdence of attanment: A comparson of methods for representng and communcatng student outcomes n systemc reform. Paper presented at the annual meetng of the Amercan Educatonal Research Assocaton, New Orleans. Mccer, T. (989). The uncorn, the normal curve, and other mprobable creatures. Psychologcal Bulletn, 05, Raudenbush, S. W. (994). Random effects models. In H. Cooper & L. V. Hedges (Eds.), The handbook of research synthess (pp. 30-3). New York: Russell Sage Foundaton. Robey, R. R. & Barckowsk, R. S. (99). Type I error and the number of teratons n Monte Carlo studes of robustness. Brtsh Journal of Mathematcal and Statstcal Psychology, 45,

Confidence Intervals for the Overall Effect Size in Random-Effects Meta-Analysis

Confidence Intervals for the Overall Effect Size in Random-Effects Meta-Analysis Psychologcal Methods 008, Vol. 13, No. 1, 31 48 Copyrght 008 by the Amercan Psychologcal Assocaton 108-989X/08/$1.00 DOI: 10.1037/108-989X.13.1.31 Confdence Intervals for the Overall Effect Sze n Random-Effects

More information

Statistics for Managers Using Microsoft Excel/SPSS Chapter 13 The Simple Linear Regression Model and Correlation

Statistics for Managers Using Microsoft Excel/SPSS Chapter 13 The Simple Linear Regression Model and Correlation Statstcs for Managers Usng Mcrosoft Excel/SPSS Chapter 13 The Smple Lnear Regresson Model and Correlaton 1999 Prentce-Hall, Inc. Chap. 13-1 Chapter Topcs Types of Regresson Models Determnng the Smple Lnear

More information

ECONOMICS 351*-A Mid-Term Exam -- Fall Term 2000 Page 1 of 13 pages. QUEEN'S UNIVERSITY AT KINGSTON Department of Economics

ECONOMICS 351*-A Mid-Term Exam -- Fall Term 2000 Page 1 of 13 pages. QUEEN'S UNIVERSITY AT KINGSTON Department of Economics ECOOMICS 35*-A Md-Term Exam -- Fall Term 000 Page of 3 pages QUEE'S UIVERSITY AT KIGSTO Department of Economcs ECOOMICS 35* - Secton A Introductory Econometrcs Fall Term 000 MID-TERM EAM ASWERS MG Abbott

More information

Negative Binomial Regression

Negative Binomial Regression STATGRAPHICS Rev. 9/16/2013 Negatve Bnomal Regresson Summary... 1 Data Input... 3 Statstcal Model... 3 Analyss Summary... 4 Analyss Optons... 7 Plot of Ftted Model... 8 Observed Versus Predcted... 10 Predctons...

More information

Statistics for Economics & Business

Statistics for Economics & Business Statstcs for Economcs & Busness Smple Lnear Regresson Learnng Objectves In ths chapter, you learn: How to use regresson analyss to predct the value of a dependent varable based on an ndependent varable

More information

Comparison of Regression Lines

Comparison of Regression Lines STATGRAPHICS Rev. 9/13/2013 Comparson of Regresson Lnes Summary... 1 Data Input... 3 Analyss Summary... 4 Plot of Ftted Model... 6 Condtonal Sums of Squares... 6 Analyss Optons... 7 Forecasts... 8 Confdence

More information

Psychology 282 Lecture #24 Outline Regression Diagnostics: Outliers

Psychology 282 Lecture #24 Outline Regression Diagnostics: Outliers Psychology 282 Lecture #24 Outlne Regresson Dagnostcs: Outlers In an earler lecture we studed the statstcal assumptons underlyng the regresson model, ncludng the followng ponts: Formal statement of assumptons.

More information

Econ107 Applied Econometrics Topic 3: Classical Model (Studenmund, Chapter 4)

Econ107 Applied Econometrics Topic 3: Classical Model (Studenmund, Chapter 4) I. Classcal Assumptons Econ7 Appled Econometrcs Topc 3: Classcal Model (Studenmund, Chapter 4) We have defned OLS and studed some algebrac propertes of OLS. In ths topc we wll study statstcal propertes

More information

Dr. Shalabh Department of Mathematics and Statistics Indian Institute of Technology Kanpur

Dr. Shalabh Department of Mathematics and Statistics Indian Institute of Technology Kanpur Analyss of Varance and Desgn of Experment-I MODULE VIII LECTURE - 34 ANALYSIS OF VARIANCE IN RANDOM-EFFECTS MODEL AND MIXED-EFFECTS EFFECTS MODEL Dr Shalabh Department of Mathematcs and Statstcs Indan

More information

Basically, if you have a dummy dependent variable you will be estimating a probability.

Basically, if you have a dummy dependent variable you will be estimating a probability. ECON 497: Lecture Notes 13 Page 1 of 1 Metropoltan State Unversty ECON 497: Research and Forecastng Lecture Notes 13 Dummy Dependent Varable Technques Studenmund Chapter 13 Bascally, f you have a dummy

More information

Chapter 13: Multiple Regression

Chapter 13: Multiple Regression Chapter 13: Multple Regresson 13.1 Developng the multple-regresson Model The general model can be descrbed as: It smplfes for two ndependent varables: The sample ft parameter b 0, b 1, and b are used to

More information

STAT 511 FINAL EXAM NAME Spring 2001

STAT 511 FINAL EXAM NAME Spring 2001 STAT 5 FINAL EXAM NAME Sprng Instructons: Ths s a closed book exam. No notes or books are allowed. ou may use a calculator but you are not allowed to store notes or formulas n the calculator. Please wrte

More information

ANSWERS CHAPTER 9. TIO 9.2: If the values are the same, the difference is 0, therefore the null hypothesis cannot be rejected.

ANSWERS CHAPTER 9. TIO 9.2: If the values are the same, the difference is 0, therefore the null hypothesis cannot be rejected. ANSWERS CHAPTER 9 THINK IT OVER thnk t over TIO 9.: χ 2 k = ( f e ) = 0 e Breakng the equaton down: the test statstc for the ch-squared dstrbuton s equal to the sum over all categores of the expected frequency

More information

2016 Wiley. Study Session 2: Ethical and Professional Standards Application

2016 Wiley. Study Session 2: Ethical and Professional Standards Application 6 Wley Study Sesson : Ethcal and Professonal Standards Applcaton LESSON : CORRECTION ANALYSIS Readng 9: Correlaton and Regresson LOS 9a: Calculate and nterpret a sample covarance and a sample correlaton

More information

A Robust Method for Calculating the Correlation Coefficient

A Robust Method for Calculating the Correlation Coefficient A Robust Method for Calculatng the Correlaton Coeffcent E.B. Nven and C. V. Deutsch Relatonshps between prmary and secondary data are frequently quantfed usng the correlaton coeffcent; however, the tradtonal

More information

1. Inference on Regression Parameters a. Finding Mean, s.d and covariance amongst estimates. 2. Confidence Intervals and Working Hotelling Bands

1. Inference on Regression Parameters a. Finding Mean, s.d and covariance amongst estimates. 2. Confidence Intervals and Working Hotelling Bands Content. Inference on Regresson Parameters a. Fndng Mean, s.d and covarance amongst estmates.. Confdence Intervals and Workng Hotellng Bands 3. Cochran s Theorem 4. General Lnear Testng 5. Measures of

More information

Chapter 9: Statistical Inference and the Relationship between Two Variables

Chapter 9: Statistical Inference and the Relationship between Two Variables Chapter 9: Statstcal Inference and the Relatonshp between Two Varables Key Words The Regresson Model The Sample Regresson Equaton The Pearson Correlaton Coeffcent Learnng Outcomes After studyng ths chapter,

More information

Assessing heterogeneity in meta-analysis: Q statistic or I2 index?

Assessing heterogeneity in meta-analysis: Q statistic or I2 index? Center for Health, Interventon, and Preventon (CHIP) CHIP Documents Unversty of Connectcut Year 006 Assessng heterogenety n meta-analyss: Q statstc or I ndex? Tana Huedo-Medna Julo Sanchez-Meca Fulgenco

More information

Simulated Power of the Discrete Cramér-von Mises Goodness-of-Fit Tests

Simulated Power of the Discrete Cramér-von Mises Goodness-of-Fit Tests Smulated of the Cramér-von Mses Goodness-of-Ft Tests Steele, M., Chaselng, J. and 3 Hurst, C. School of Mathematcal and Physcal Scences, James Cook Unversty, Australan School of Envronmental Studes, Grffth

More information

Computation of Higher Order Moments from Two Multinomial Overdispersion Likelihood Models

Computation of Higher Order Moments from Two Multinomial Overdispersion Likelihood Models Computaton of Hgher Order Moments from Two Multnomal Overdsperson Lkelhood Models BY J. T. NEWCOMER, N. K. NEERCHAL Department of Mathematcs and Statstcs, Unversty of Maryland, Baltmore County, Baltmore,

More information

Chapter 8 Indicator Variables

Chapter 8 Indicator Variables Chapter 8 Indcator Varables In general, e explanatory varables n any regresson analyss are assumed to be quanttatve n nature. For example, e varables lke temperature, dstance, age etc. are quanttatve n

More information

Testing for seasonal unit roots in heterogeneous panels

Testing for seasonal unit roots in heterogeneous panels Testng for seasonal unt roots n heterogeneous panels Jesus Otero * Facultad de Economía Unversdad del Rosaro, Colomba Jeremy Smth Department of Economcs Unversty of arwck Monca Gulett Aston Busness School

More information

LOGIT ANALYSIS. A.K. VASISHT Indian Agricultural Statistics Research Institute, Library Avenue, New Delhi

LOGIT ANALYSIS. A.K. VASISHT Indian Agricultural Statistics Research Institute, Library Avenue, New Delhi LOGIT ANALYSIS A.K. VASISHT Indan Agrcultural Statstcs Research Insttute, Lbrary Avenue, New Delh-0 02 amtvassht@asr.res.n. Introducton In dummy regresson varable models, t s assumed mplctly that the dependent

More information

Interval Estimation in the Classical Normal Linear Regression Model. 1. Introduction

Interval Estimation in the Classical Normal Linear Regression Model. 1. Introduction ECONOMICS 35* -- NOTE 7 ECON 35* -- NOTE 7 Interval Estmaton n the Classcal Normal Lnear Regresson Model Ths note outlnes the basc elements of nterval estmaton n the Classcal Normal Lnear Regresson Model

More information

Economics 130. Lecture 4 Simple Linear Regression Continued

Economics 130. Lecture 4 Simple Linear Regression Continued Economcs 130 Lecture 4 Contnued Readngs for Week 4 Text, Chapter and 3. We contnue wth addressng our second ssue + add n how we evaluate these relatonshps: Where do we get data to do ths analyss? How do

More information

Chapter 11: Simple Linear Regression and Correlation

Chapter 11: Simple Linear Regression and Correlation Chapter 11: Smple Lnear Regresson and Correlaton 11-1 Emprcal Models 11-2 Smple Lnear Regresson 11-3 Propertes of the Least Squares Estmators 11-4 Hypothess Test n Smple Lnear Regresson 11-4.1 Use of t-tests

More information

Department of Quantitative Methods & Information Systems. Time Series and Their Components QMIS 320. Chapter 6

Department of Quantitative Methods & Information Systems. Time Series and Their Components QMIS 320. Chapter 6 Department of Quanttatve Methods & Informaton Systems Tme Seres and Ther Components QMIS 30 Chapter 6 Fall 00 Dr. Mohammad Zanal These sldes were modfed from ther orgnal source for educatonal purpose only.

More information

4 Analysis of Variance (ANOVA) 5 ANOVA. 5.1 Introduction. 5.2 Fixed Effects ANOVA

4 Analysis of Variance (ANOVA) 5 ANOVA. 5.1 Introduction. 5.2 Fixed Effects ANOVA 4 Analyss of Varance (ANOVA) 5 ANOVA 51 Introducton ANOVA ANOVA s a way to estmate and test the means of multple populatons We wll start wth one-way ANOVA If the populatons ncluded n the study are selected

More information

Topic- 11 The Analysis of Variance

Topic- 11 The Analysis of Variance Topc- 11 The Analyss of Varance Expermental Desgn The samplng plan or expermental desgn determnes the way that a sample s selected. In an observatonal study, the expermenter observes data that already

More information

ANOMALIES OF THE MAGNITUDE OF THE BIAS OF THE MAXIMUM LIKELIHOOD ESTIMATOR OF THE REGRESSION SLOPE

ANOMALIES OF THE MAGNITUDE OF THE BIAS OF THE MAXIMUM LIKELIHOOD ESTIMATOR OF THE REGRESSION SLOPE P a g e ANOMALIES OF THE MAGNITUDE OF THE BIAS OF THE MAXIMUM LIKELIHOOD ESTIMATOR OF THE REGRESSION SLOPE Darmud O Drscoll ¹, Donald E. Ramrez ² ¹ Head of Department of Mathematcs and Computer Studes

More information

/ n ) are compared. The logic is: if the two

/ n ) are compared. The logic is: if the two STAT C141, Sprng 2005 Lecture 13 Two sample tests One sample tests: examples of goodness of ft tests, where we are testng whether our data supports predctons. Two sample tests: called as tests of ndependence

More information

Non-Mixture Cure Model for Interval Censored Data: Simulation Study ABSTRACT

Non-Mixture Cure Model for Interval Censored Data: Simulation Study ABSTRACT Malaysan Journal of Mathematcal Scences 8(S): 37-44 (2014) Specal Issue: Internatonal Conference on Mathematcal Scences and Statstcs 2013 (ICMSS2013) MALAYSIAN JOURNAL OF MATHEMATICAL SCIENCES Journal

More information

A Comparative Study for Estimation Parameters in Panel Data Model

A Comparative Study for Estimation Parameters in Panel Data Model A Comparatve Study for Estmaton Parameters n Panel Data Model Ahmed H. Youssef and Mohamed R. Abonazel hs paper examnes the panel data models when the regresson coeffcents are fxed random and mxed and

More information

See Book Chapter 11 2 nd Edition (Chapter 10 1 st Edition)

See Book Chapter 11 2 nd Edition (Chapter 10 1 st Edition) Count Data Models See Book Chapter 11 2 nd Edton (Chapter 10 1 st Edton) Count data consst of non-negatve nteger values Examples: number of drver route changes per week, the number of trp departure changes

More information

DERIVATION OF THE PROBABILITY PLOT CORRELATION COEFFICIENT TEST STATISTICS FOR THE GENERALIZED LOGISTIC DISTRIBUTION

DERIVATION OF THE PROBABILITY PLOT CORRELATION COEFFICIENT TEST STATISTICS FOR THE GENERALIZED LOGISTIC DISTRIBUTION Internatonal Worshop ADVANCES IN STATISTICAL HYDROLOGY May 3-5, Taormna, Italy DERIVATION OF THE PROBABILITY PLOT CORRELATION COEFFICIENT TEST STATISTICS FOR THE GENERALIZED LOGISTIC DISTRIBUTION by Sooyoung

More information

RESAMPLING TESTS FOR META-ANALYSIS OF ECOLOGICAL DATA

RESAMPLING TESTS FOR META-ANALYSIS OF ECOLOGICAL DATA June 997 REPORTS 277 Ecology, 78(5), 997, pp. 277 283 997 by the Ecologcal Socety of Amerca RESAMPLING TESTS FOR META-ANALYSIS OF ECOLOGICAL DATA DEAN C. ADAMS, JESSICA GUREVITCH, AND MICHAEL S. ROSENBERG

More information

Basic Business Statistics, 10/e

Basic Business Statistics, 10/e Chapter 13 13-1 Basc Busness Statstcs 11 th Edton Chapter 13 Smple Lnear Regresson Basc Busness Statstcs, 11e 009 Prentce-Hall, Inc. Chap 13-1 Learnng Objectves In ths chapter, you learn: How to use regresson

More information

is the calculated value of the dependent variable at point i. The best parameters have values that minimize the squares of the errors

is the calculated value of the dependent variable at point i. The best parameters have values that minimize the squares of the errors Multple Lnear and Polynomal Regresson wth Statstcal Analyss Gven a set of data of measured (or observed) values of a dependent varable: y versus n ndependent varables x 1, x, x n, multple lnear regresson

More information

Linear Approximation with Regularization and Moving Least Squares

Linear Approximation with Regularization and Moving Least Squares Lnear Approxmaton wth Regularzaton and Movng Least Squares Igor Grešovn May 007 Revson 4.6 (Revson : March 004). 5 4 3 0.5 3 3.5 4 Contents: Lnear Fttng...4. Weghted Least Squares n Functon Approxmaton...

More information

Statistical tables are provided Two Hours UNIVERSITY OF MANCHESTER. Date: Wednesday 4 th June 2008 Time: 1400 to 1600

Statistical tables are provided Two Hours UNIVERSITY OF MANCHESTER. Date: Wednesday 4 th June 2008 Time: 1400 to 1600 Statstcal tables are provded Two Hours UNIVERSITY OF MNCHESTER Medcal Statstcs Date: Wednesday 4 th June 008 Tme: 1400 to 1600 MT3807 Electronc calculators may be used provded that they conform to Unversty

More information

Chapter 14 Simple Linear Regression

Chapter 14 Simple Linear Regression Chapter 4 Smple Lnear Regresson Chapter 4 - Smple Lnear Regresson Manageral decsons often are based on the relatonshp between two or more varables. Regresson analss can be used to develop an equaton showng

More information

Effective plots to assess bias and precision in method comparison studies

Effective plots to assess bias and precision in method comparison studies Effectve plots to assess bas and precson n method comparson studes Bern, November, 016 Patrck Taffé, PhD Insttute of Socal and Preventve Medcne () Unversty of Lausanne, Swtzerland Patrck.Taffe@chuv.ch

More information

Comparison of the Population Variance Estimators. of 2-Parameter Exponential Distribution Based on. Multiple Criteria Decision Making Method

Comparison of the Population Variance Estimators. of 2-Parameter Exponential Distribution Based on. Multiple Criteria Decision Making Method Appled Mathematcal Scences, Vol. 7, 0, no. 47, 07-0 HIARI Ltd, www.m-hkar.com Comparson of the Populaton Varance Estmators of -Parameter Exponental Dstrbuton Based on Multple Crtera Decson Makng Method

More information

LINEAR REGRESSION ANALYSIS. MODULE VIII Lecture Indicator Variables

LINEAR REGRESSION ANALYSIS. MODULE VIII Lecture Indicator Variables LINEAR REGRESSION ANALYSIS MODULE VIII Lecture - 7 Indcator Varables Dr. Shalabh Department of Maematcs and Statstcs Indan Insttute of Technology Kanpur Indcator varables versus quanttatve explanatory

More information

UNIVERSITY OF TORONTO Faculty of Arts and Science. December 2005 Examinations STA437H1F/STA1005HF. Duration - 3 hours

UNIVERSITY OF TORONTO Faculty of Arts and Science. December 2005 Examinations STA437H1F/STA1005HF. Duration - 3 hours UNIVERSITY OF TORONTO Faculty of Arts and Scence December 005 Examnatons STA47HF/STA005HF Duraton - hours AIDS ALLOWED: (to be suppled by the student) Non-programmable calculator One handwrtten 8.5'' x

More information

DrPH Seminar Session 3. Quantitative Synthesis. Qualitative Synthesis e.g., GRADE

DrPH Seminar Session 3. Quantitative Synthesis. Qualitative Synthesis e.g., GRADE DrPH Semnar Sesson 3 Quanttatve Synthess Focusng on Heterogenety Qualtatve Synthess e.g., GRADE Me Chung, PhD, MPH Research Assstant Professor Nutrton/Infecton Unt, Department of Publc Health and Communty

More information

x = , so that calculated

x = , so that calculated Stat 4, secton Sngle Factor ANOVA notes by Tm Plachowsk n chapter 8 we conducted hypothess tests n whch we compared a sngle sample s mean or proporton to some hypotheszed value Chapter 9 expanded ths to

More information

Department of Statistics University of Toronto STA305H1S / 1004 HS Design and Analysis of Experiments Term Test - Winter Solution

Department of Statistics University of Toronto STA305H1S / 1004 HS Design and Analysis of Experiments Term Test - Winter Solution Department of Statstcs Unversty of Toronto STA35HS / HS Desgn and Analyss of Experments Term Test - Wnter - Soluton February, Last Name: Frst Name: Student Number: Instructons: Tme: hours. Ads: a non-programmable

More information

Durban Watson for Testing the Lack-of-Fit of Polynomial Regression Models without Replications

Durban Watson for Testing the Lack-of-Fit of Polynomial Regression Models without Replications Durban Watson for Testng the Lack-of-Ft of Polynomal Regresson Models wthout Replcatons Ruba A. Alyaf, Maha A. Omar, Abdullah A. Al-Shha ralyaf@ksu.edu.sa, maomar@ksu.edu.sa, aalshha@ksu.edu.sa Department

More information

Meta-Analysis What is it? Why is it important? How do you do it? What is meta-analysis? Good books on meta-analysis

Meta-Analysis What is it? Why is it important? How do you do it? What is meta-analysis? Good books on meta-analysis Meta-Analyss What s t? Why s t mportant? How do you do t? (Summer) What s meta-analyss? Meta-analyss can be thought of as a form of survey research n whch research reports are the unts surveyed (Lpsey

More information

Sampling Theory MODULE VII LECTURE - 23 VARYING PROBABILITY SAMPLING

Sampling Theory MODULE VII LECTURE - 23 VARYING PROBABILITY SAMPLING Samplng heory MODULE VII LECURE - 3 VARYIG PROBABILIY SAMPLIG DR. SHALABH DEPARME OF MAHEMAICS AD SAISICS IDIA ISIUE OF ECHOLOGY KAPUR he smple random samplng scheme provdes a random sample where every

More information

Uncertainty as the Overlap of Alternate Conditional Distributions

Uncertainty as the Overlap of Alternate Conditional Distributions Uncertanty as the Overlap of Alternate Condtonal Dstrbutons Olena Babak and Clayton V. Deutsch Centre for Computatonal Geostatstcs Department of Cvl & Envronmental Engneerng Unversty of Alberta An mportant

More information

Statistical Evaluation of WATFLOOD

Statistical Evaluation of WATFLOOD tatstcal Evaluaton of WATFLD By: Angela MacLean, Dept. of Cvl & Envronmental Engneerng, Unversty of Waterloo, n. ctober, 005 The statstcs program assocated wth WATFLD uses spl.csv fle that s produced wth

More information

Here is the rationale: If X and y have a strong positive relationship to one another, then ( x x) will tend to be positive when ( y y)

Here is the rationale: If X and y have a strong positive relationship to one another, then ( x x) will tend to be positive when ( y y) Secton 1.5 Correlaton In the prevous sectons, we looked at regresson and the value r was a measurement of how much of the varaton n y can be attrbuted to the lnear relatonshp between y and x. In ths secton,

More information

Statistics for Business and Economics

Statistics for Business and Economics Statstcs for Busness and Economcs Chapter 11 Smple Regresson Copyrght 010 Pearson Educaton, Inc. Publshng as Prentce Hall Ch. 11-1 11.1 Overvew of Lnear Models n An equaton can be ft to show the best lnear

More information

USE OF DOUBLE SAMPLING SCHEME IN ESTIMATING THE MEAN OF STRATIFIED POPULATION UNDER NON-RESPONSE

USE OF DOUBLE SAMPLING SCHEME IN ESTIMATING THE MEAN OF STRATIFIED POPULATION UNDER NON-RESPONSE STATISTICA, anno LXXV, n. 4, 015 USE OF DOUBLE SAMPLING SCHEME IN ESTIMATING THE MEAN OF STRATIFIED POPULATION UNDER NON-RESPONSE Manoj K. Chaudhary 1 Department of Statstcs, Banaras Hndu Unversty, Varanas,

More information

Parametric fractional imputation for missing data analysis. Jae Kwang Kim Survey Working Group Seminar March 29, 2010

Parametric fractional imputation for missing data analysis. Jae Kwang Kim Survey Working Group Seminar March 29, 2010 Parametrc fractonal mputaton for mssng data analyss Jae Kwang Km Survey Workng Group Semnar March 29, 2010 1 Outlne Introducton Proposed method Fractonal mputaton Approxmaton Varance estmaton Multple mputaton

More information

DO NOT OPEN THE QUESTION PAPER UNTIL INSTRUCTED TO DO SO BY THE CHIEF INVIGILATOR. Introductory Econometrics 1 hour 30 minutes

DO NOT OPEN THE QUESTION PAPER UNTIL INSTRUCTED TO DO SO BY THE CHIEF INVIGILATOR. Introductory Econometrics 1 hour 30 minutes 25/6 Canddates Only January Examnatons 26 Student Number: Desk Number:...... DO NOT OPEN THE QUESTION PAPER UNTIL INSTRUCTED TO DO SO BY THE CHIEF INVIGILATOR Department Module Code Module Ttle Exam Duraton

More information

Chapter 2 - The Simple Linear Regression Model S =0. e i is a random error. S β2 β. This is a minimization problem. Solution is a calculus exercise.

Chapter 2 - The Simple Linear Regression Model S =0. e i is a random error. S β2 β. This is a minimization problem. Solution is a calculus exercise. Chapter - The Smple Lnear Regresson Model The lnear regresson equaton s: where y + = β + β e for =,..., y and are observable varables e s a random error How can an estmaton rule be constructed for the

More information

BOOTSTRAP METHOD FOR TESTING OF EQUALITY OF SEVERAL MEANS. M. Krishna Reddy, B. Naveen Kumar and Y. Ramu

BOOTSTRAP METHOD FOR TESTING OF EQUALITY OF SEVERAL MEANS. M. Krishna Reddy, B. Naveen Kumar and Y. Ramu BOOTSTRAP METHOD FOR TESTING OF EQUALITY OF SEVERAL MEANS M. Krshna Reddy, B. Naveen Kumar and Y. Ramu Department of Statstcs, Osmana Unversty, Hyderabad -500 007, Inda. nanbyrozu@gmal.com, ramu0@gmal.com

More information

Lecture 4 Hypothesis Testing

Lecture 4 Hypothesis Testing Lecture 4 Hypothess Testng We may wsh to test pror hypotheses about the coeffcents we estmate. We can use the estmates to test whether the data rejects our hypothess. An example mght be that we wsh to

More information

Jon Deeks and Julian Higgins. on Behalf of the Statistical Methods Group of The Cochrane Collaboration. April 2005

Jon Deeks and Julian Higgins. on Behalf of the Statistical Methods Group of The Cochrane Collaboration. April 2005 Standard statstcal algorthms n Cochrane revews Verson 5 Jon Deeks and Julan Hggns on Behalf of the Statstcal Methods Group of The Cochrane Collaboraton Aprl 005 Data structure Consder a meta-analyss of

More information

Global Sensitivity. Tuesday 20 th February, 2018

Global Sensitivity. Tuesday 20 th February, 2018 Global Senstvty Tuesday 2 th February, 28 ) Local Senstvty Most senstvty analyses [] are based on local estmates of senstvty, typcally by expandng the response n a Taylor seres about some specfc values

More information

An accurate test for homogeneity of odds ratios based on Cochran s Q-statistic

An accurate test for homogeneity of odds ratios based on Cochran s Q-statistic Kulnskaya and Dollnger BMC Medcal Research Methodology (2015) 15:49 DOI 10.1186/s12874-015-0034-x TECHNICAL ADVANCE Open Access An accurate test for homogenety of odds ratos based on Cochran s Q-statstc

More information

Chapter 3 Describing Data Using Numerical Measures

Chapter 3 Describing Data Using Numerical Measures Chapter 3 Student Lecture Notes 3-1 Chapter 3 Descrbng Data Usng Numercal Measures Fall 2006 Fundamentals of Busness Statstcs 1 Chapter Goals To establsh the usefulness of summary measures of data. The

More information

Lecture 3 Stat102, Spring 2007

Lecture 3 Stat102, Spring 2007 Lecture 3 Stat0, Sprng 007 Chapter 3. 3.: Introducton to regresson analyss Lnear regresson as a descrptve technque The least-squares equatons Chapter 3.3 Samplng dstrbuton of b 0, b. Contnued n net lecture

More information

Computing MLE Bias Empirically

Computing MLE Bias Empirically Computng MLE Bas Emprcally Kar Wa Lm Australan atonal Unversty January 3, 27 Abstract Ths note studes the bas arses from the MLE estmate of the rate parameter and the mean parameter of an exponental dstrbuton.

More information

STATISTICS QUESTIONS. Step by Step Solutions.

STATISTICS QUESTIONS. Step by Step Solutions. STATISTICS QUESTIONS Step by Step Solutons www.mathcracker.com 9//016 Problem 1: A researcher s nterested n the effects of famly sze on delnquency for a group of offenders and examnes famles wth one to

More information

January Examinations 2015

January Examinations 2015 24/5 Canddates Only January Examnatons 25 DO NOT OPEN THE QUESTION PAPER UNTIL INSTRUCTED TO DO SO BY THE CHIEF INVIGILATOR STUDENT CANDIDATE NO.. Department Module Code Module Ttle Exam Duraton (n words)

More information

Dr. Shalabh Department of Mathematics and Statistics Indian Institute of Technology Kanpur

Dr. Shalabh Department of Mathematics and Statistics Indian Institute of Technology Kanpur Analyss of Varance and Desgn of Experment-I MODULE VII LECTURE - 3 ANALYSIS OF COVARIANCE Dr Shalabh Department of Mathematcs and Statstcs Indan Insttute of Technology Kanpur Any scentfc experment s performed

More information

Problem of Estimation. Ordinary Least Squares (OLS) Ordinary Least Squares Method. Basic Econometrics in Transportation. Bivariate Regression Analysis

Problem of Estimation. Ordinary Least Squares (OLS) Ordinary Least Squares Method. Basic Econometrics in Transportation. Bivariate Regression Analysis 1/60 Problem of Estmaton Basc Econometrcs n Transportaton Bvarate Regresson Analyss Amr Samm Cvl Engneerng Department Sharf Unversty of Technology Ordnary Least Squares (OLS) Maxmum Lkelhood (ML) Generally,

More information

A nonparametric two-sample wald test of equality of variances

A nonparametric two-sample wald test of equality of variances Unversty of Wollongong Research Onlne Centre for Statstcal & Survey Methodology Workng Paper Seres Faculty of Engneerng and Informaton Scences 0 A nonparametrc two-sample wald test of equalty of varances

More information

Assessment of Site Amplification Effect from Input Energy Spectra of Strong Ground Motion

Assessment of Site Amplification Effect from Input Energy Spectra of Strong Ground Motion Assessment of Ste Amplfcaton Effect from Input Energy Spectra of Strong Ground Moton M.S. Gong & L.L Xe Key Laboratory of Earthquake Engneerng and Engneerng Vbraton,Insttute of Engneerng Mechancs, CEA,

More information

The Ordinary Least Squares (OLS) Estimator

The Ordinary Least Squares (OLS) Estimator The Ordnary Least Squares (OLS) Estmator 1 Regresson Analyss Regresson Analyss: a statstcal technque for nvestgatng and modelng the relatonshp between varables. Applcatons: Engneerng, the physcal and chemcal

More information

ERROR RATES STABILITY OF THE HOMOSCEDASTIC DISCRIMINANT FUNCTION

ERROR RATES STABILITY OF THE HOMOSCEDASTIC DISCRIMINANT FUNCTION ISSN - 77-0593 UNAAB 00 Journal of Natural Scences, Engneerng and Technology ERROR RATES STABILITY OF THE HOMOSCEDASTIC DISCRIMINANT FUNCTION A. ADEBANJI, S. NOKOE AND O. IYANIWURA 3 *Department of Mathematcs,

More information

Chapter 5 Multilevel Models

Chapter 5 Multilevel Models Chapter 5 Multlevel Models 5.1 Cross-sectonal multlevel models 5.1.1 Two-level models 5.1.2 Multple level models 5.1.3 Multple level modelng n other felds 5.2 Longtudnal multlevel models 5.2.1 Two-level

More information

COMPARISON OF SOME RELIABILITY CHARACTERISTICS BETWEEN REDUNDANT SYSTEMS REQUIRING SUPPORTING UNITS FOR THEIR OPERATIONS

COMPARISON OF SOME RELIABILITY CHARACTERISTICS BETWEEN REDUNDANT SYSTEMS REQUIRING SUPPORTING UNITS FOR THEIR OPERATIONS Avalable onlne at http://sck.org J. Math. Comput. Sc. 3 (3), No., 6-3 ISSN: 97-537 COMPARISON OF SOME RELIABILITY CHARACTERISTICS BETWEEN REDUNDANT SYSTEMS REQUIRING SUPPORTING UNITS FOR THEIR OPERATIONS

More information

Andreas C. Drichoutis Agriculural University of Athens. Abstract

Andreas C. Drichoutis Agriculural University of Athens. Abstract Heteroskedastcty, the sngle crossng property and ordered response models Andreas C. Drchouts Agrculural Unversty of Athens Panagots Lazards Agrculural Unversty of Athens Rodolfo M. Nayga, Jr. Texas AMUnversty

More information

A Monte Carlo Study for Swamy s Estimate of Random Coefficient Panel Data Model

A Monte Carlo Study for Swamy s Estimate of Random Coefficient Panel Data Model A Monte Carlo Study for Swamy s Estmate of Random Coeffcent Panel Data Model Aman Mousa, Ahmed H. Youssef and Mohamed R. Abonazel Department of Appled Statstcs and Econometrcs, Instute of Statstcal Studes

More information

SIMPLE LINEAR REGRESSION

SIMPLE LINEAR REGRESSION Smple Lnear Regresson and Correlaton Introducton Prevousl, our attenton has been focused on one varable whch we desgnated b x. Frequentl, t s desrable to learn somethng about the relatonshp between two

More information

Efficient nonresponse weighting adjustment using estimated response probability

Efficient nonresponse weighting adjustment using estimated response probability Effcent nonresponse weghtng adjustment usng estmated response probablty Jae Kwang Km Department of Appled Statstcs, Yonse Unversty, Seoul, 120-749, KOREA Key Words: Regresson estmator, Propensty score,

More information

ANSWERS. Problem 1. and the moment generating function (mgf) by. defined for any real t. Use this to show that E( U) var( U)

ANSWERS. Problem 1. and the moment generating function (mgf) by. defined for any real t. Use this to show that E( U) var( U) Econ 413 Exam 13 H ANSWERS Settet er nndelt 9 deloppgaver, A,B,C, som alle anbefales å telle lkt for å gøre det ltt lettere å stå. Svar er gtt . Unfortunately, there s a prntng error n the hnt of

More information

[The following data appear in Wooldridge Q2.3.] The table below contains the ACT score and college GPA for eight college students.

[The following data appear in Wooldridge Q2.3.] The table below contains the ACT score and college GPA for eight college students. PPOL 59-3 Problem Set Exercses n Smple Regresson Due n class /8/7 In ths problem set, you are asked to compute varous statstcs by hand to gve you a better sense of the mechancs of the Pearson correlaton

More information

STAT 3008 Applied Regression Analysis

STAT 3008 Applied Regression Analysis STAT 3008 Appled Regresson Analyss Tutoral : Smple Lnear Regresson LAI Chun He Department of Statstcs, The Chnese Unversty of Hong Kong 1 Model Assumpton To quantfy the relatonshp between two factors,

More information

Appendix B: Resampling Algorithms

Appendix B: Resampling Algorithms 407 Appendx B: Resamplng Algorthms A common problem of all partcle flters s the degeneracy of weghts, whch conssts of the unbounded ncrease of the varance of the mportance weghts ω [ ] of the partcles

More information

ASYMPTOTIC PROPERTIES OF ESTIMATES FOR THE PARAMETERS IN THE LOGISTIC REGRESSION MODEL

ASYMPTOTIC PROPERTIES OF ESTIMATES FOR THE PARAMETERS IN THE LOGISTIC REGRESSION MODEL Asymptotc Asan-Afrcan Propertes Journal of Estmates Economcs for and the Econometrcs, Parameters n Vol. the Logstc, No., Regresson 20: 65-74 Model 65 ASYMPTOTIC PROPERTIES OF ESTIMATES FOR THE PARAMETERS

More information

Test for Intraclass Correlation Coefficient under Unequal Family Sizes

Test for Intraclass Correlation Coefficient under Unequal Family Sizes Journal of Modern Appled Statstcal Methods Volume Issue Artcle 9 --03 Test for Intraclass Correlaton Coeffcent under Unequal Famly Szes Madhusudan Bhandary Columbus State Unversty, Columbus, GA, bhandary_madhusudan@colstate.edu

More information

UCLA STAT 13 Introduction to Statistical Methods for the Life and Health Sciences. Chapter 11 Analysis of Variance - ANOVA. Instructor: Ivo Dinov,

UCLA STAT 13 Introduction to Statistical Methods for the Life and Health Sciences. Chapter 11 Analysis of Variance - ANOVA. Instructor: Ivo Dinov, UCLA STAT 3 ntroducton to Statstcal Methods for the Lfe and Health Scences nstructor: vo Dnov, Asst. Prof. of Statstcs and Neurology Chapter Analyss of Varance - ANOVA Teachng Assstants: Fred Phoa, Anwer

More information

CHAPTER 8. Exercise Solutions

CHAPTER 8. Exercise Solutions CHAPTER 8 Exercse Solutons 77 Chapter 8, Exercse Solutons, Prncples of Econometrcs, 3e 78 EXERCISE 8. When = N N N ( x x) ( x x) ( x x) = = = N = = = N N N ( x ) ( ) ( ) ( x x ) x x x x x = = = = Chapter

More information

Statistics II Final Exam 26/6/18

Statistics II Final Exam 26/6/18 Statstcs II Fnal Exam 26/6/18 Academc Year 2017/18 Solutons Exam duraton: 2 h 30 mn 1. (3 ponts) A town hall s conductng a study to determne the amount of leftover food produced by the restaurants n the

More information

Econ107 Applied Econometrics Topic 9: Heteroskedasticity (Studenmund, Chapter 10)

Econ107 Applied Econometrics Topic 9: Heteroskedasticity (Studenmund, Chapter 10) I. Defnton and Problems Econ7 Appled Econometrcs Topc 9: Heteroskedastcty (Studenmund, Chapter ) We now relax another classcal assumpton. Ths s a problem that arses often wth cross sectons of ndvduals,

More information

THE EFFECTS OF NON-NORMALITY ON TYPE III ERROR FOR COMPARING INDEPENDENT MEANS

THE EFFECTS OF NON-NORMALITY ON TYPE III ERROR FOR COMPARING INDEPENDENT MEANS THE EFFECTS OF NON-NORMALITY ON TYPE III ERROR FOR COMPARING INDEPENDENT MEANS Mehmet MENDES PhD, Assocate Professor, Canaale Onsez Mart Unversty, Agrculture Faculty, Anmal Scence Department, Bometry and

More information

Modeling and Simulation NETW 707

Modeling and Simulation NETW 707 Modelng and Smulaton NETW 707 Lecture 5 Tests for Random Numbers Course Instructor: Dr.-Ing. Magge Mashaly magge.ezzat@guc.edu.eg C3.220 1 Propertes of Random Numbers Random Number Generators (RNGs) must

More information

LINEAR REGRESSION ANALYSIS. MODULE IX Lecture Multicollinearity

LINEAR REGRESSION ANALYSIS. MODULE IX Lecture Multicollinearity LINEAR REGRESSION ANALYSIS MODULE IX Lecture - 31 Multcollnearty Dr. Shalabh Department of Mathematcs and Statstcs Indan Insttute of Technology Kanpur 6. Rdge regresson The OLSE s the best lnear unbased

More information

UNR Joint Economics Working Paper Series Working Paper No Further Analysis of the Zipf Law: Does the Rank-Size Rule Really Exist?

UNR Joint Economics Working Paper Series Working Paper No Further Analysis of the Zipf Law: Does the Rank-Size Rule Really Exist? UNR Jont Economcs Workng Paper Seres Workng Paper No. 08-005 Further Analyss of the Zpf Law: Does the Rank-Sze Rule Really Exst? Fungsa Nota and Shunfeng Song Department of Economcs /030 Unversty of Nevada,

More information

e i is a random error

e i is a random error Chapter - The Smple Lnear Regresson Model The lnear regresson equaton s: where + β + β e for,..., and are observable varables e s a random error How can an estmaton rule be constructed for the unknown

More information

The Multiple Classical Linear Regression Model (CLRM): Specification and Assumptions. 1. Introduction

The Multiple Classical Linear Regression Model (CLRM): Specification and Assumptions. 1. Introduction ECONOMICS 5* -- NOTE (Summary) ECON 5* -- NOTE The Multple Classcal Lnear Regresson Model (CLRM): Specfcaton and Assumptons. Introducton CLRM stands for the Classcal Lnear Regresson Model. The CLRM s also

More information

Predictive Analytics : QM901.1x Prof U Dinesh Kumar, IIMB. All Rights Reserved, Indian Institute of Management Bangalore

Predictive Analytics : QM901.1x Prof U Dinesh Kumar, IIMB. All Rights Reserved, Indian Institute of Management Bangalore Sesson Outlne Introducton to classfcaton problems and dscrete choce models. Introducton to Logstcs Regresson. Logstc functon and Logt functon. Maxmum Lkelhood Estmator (MLE) for estmaton of LR parameters.

More information

Copyright 2017 by Taylor Enterprises, Inc., All Rights Reserved. Adjusted Control Limits for P Charts. Dr. Wayne A. Taylor

Copyright 2017 by Taylor Enterprises, Inc., All Rights Reserved. Adjusted Control Limits for P Charts. Dr. Wayne A. Taylor Taylor Enterprses, Inc. Control Lmts for P Charts Copyrght 2017 by Taylor Enterprses, Inc., All Rghts Reserved. Control Lmts for P Charts Dr. Wayne A. Taylor Abstract: P charts are used for count data

More information

Issues To Consider when Estimating Health Care Costs with Generalized Linear Models (GLMs): To Gamma/Log Or Not To Gamma/Log? That Is The New Question

Issues To Consider when Estimating Health Care Costs with Generalized Linear Models (GLMs): To Gamma/Log Or Not To Gamma/Log? That Is The New Question Issues To Consder when Estmatng Health Care Costs wth Generalzed Lnear Models (GLMs): To Gamma/Log Or Not To Gamma/Log? That Is The New Queston ISPOR 20th Annual Internatonal Meetng May 19, 2015 Jalpa

More information