A Monte-Carlo study of asymptotically robust tests for correlation coefficients

Size: px
Start display at page:

Download "A Monte-Carlo study of asymptotically robust tests for correlation coefficients"

Transcription

1 Biometrika (1973), 6, 3, p Printed in Great Britain A Monte-Carlo study of asymptotically robust tests for correlation coefficients BY G. T. DUNCAN AND M. W. J. LAYAKD University of California, Davis SUMMABY Monte-Carlo simulation is used to compare the small-sample performance of the usual normal theory procedures for inference about correlation coefficients with that of two asymptotically robust procedures, one of which is based on a grouping of the observations and the other on the jackknife technique. The sampled distributions comprise the normal and five nonnormal distributions. The small-sample results support the conclusion based on asymptotic theory that the normal test is not robust. The jackknife procedure works well for most of the sampled distributions. Some key words: Correlation coefficients; Hypothesis testing; Confidence intervals; Robustness; Large sample theory;. 1. INTRODUCTION Tests of hypotheses and confidence intervals concerning correlation coefficients in bivariate normal populations are commonly based on Fisher's z transformation of the sample correlation coefficient, tanh" 1 r. This statistic is taken to be approximately normally distributed with mean t&nh~ 1 p and variance 1/(TI 3), the accuracy of the approximation being adequate for moderately small samples and improving as n increases. For a bivariate population, not necessarily normal but having finite fourth moments, it can be shown that ^/n(tanh~ 1 r tanh" 1^) converges in distribution to N{,cr 2 {p)}, where The A's are the standardized cumulants of order four of the bivariate distribution; for example, A^ = KWI "* "?)' where a% and crj are the variances of the marginal distributions. Details are given in an unpublished Stanford University report by M. W. J. Layard. Note that A^ and A. M are the kurtoses of the marginal distributions. If the distribution is bivariate normal, all the A's vanish, and o" 2 (p) = 1. If the components are independent, we have p = and A 22 =, and - 2 (O) = 1. Hence the normal theory test for independence based on tanh" 1 r is asymptotically valid for any population having finite fourth moments. However, if independence does not hold the asymptotic variance of tanh -1 r is not in general 1/n, whether or not p =, and so normal theory procedures about p based on tanh" 1 r may not be valid, even asymptotically, if the population is not in fact bivariate normal. The effects of nonnormality on the distribution of r were first investigated by Baker (193) and Pearson (1931, 1932). An interesting historical survey of this and succeeding work is given by Kowalskd (1972). Correlation coefficients are often used by research workers even though their data are nonnormal. This being a subject of controversy (Nefzger & Drasgow, (1-1) Downloaded from at Penn State University (Paterno Lib) on May 9, 216

2 552 G. T. DUNCAN AND M. W. J. LAYABD 1957; Norris & Heljm, 1961), we present inference procedures which are asymptotically robust against nonnormality and investigate their performance for samples of moderate size. By asymptotically robust we mean that the size of a test, or the coverage of a confidence interval, converges to the nominal level as the sample size tends to infinity. Naturally, in spite of its wide use, it can be argued that correlation has limitations as opposed to regression in examining the relationship between two variables. An estimated regression line, # = &+fix, emphasizes the linear nature of the association, gives a pictorial representation of the relationship, and explicitly allows prediction. In bivariate normal regression analysis one would use t = r N /(n-2)/^/(l -r 2 ) to test independence, i.e. /? =, rather than Fisher's z, and of course the same statistic is commonly used to test the equivalent hypothesis p =. However, one would anticipate the same problems of nonrobustness in the nonindependent case with t as with z. It can be shown that the asymptotic variance of t when /? = is 1 + A M, identical to that of z*jn. Asymptotic nonrobustness can also be demonstrated for the t test of the hypothesis /? = /? +. The asymptotically robust procedures discussed here can readily be adapted to problems of inference about the regression coefficient /?. We now describe the proposed procedures in the context of the two-sample problem of testing the equality of correlation coefficients; the procedures are easily modified to the one-sample problem. Suppose that we have independent samples (X^, F (1 ),..., (X^^, T^^) (t = 1,2) from bivariate populations with distribution functions F and O, correlations p x and /o a, and finite fourth moments. We wish to test where, rj, <r and T are unspecified constants, versus H A : p x 4= p 2. The hypothesis HQ implies Pi = Pi but in general the reverse implication does not hold, though it does in the normal case. The choice of H o rather than the more general hypothesis p x = p t ensures that the fourth-order standardized cumulants of the two distributions are equal. The first test, which we designate the test, is analogous to the procedure suggested by (1953) for testing homogeneity of variances. Divide the tth sample randomly into J { subsamples of size m, supposing that n { = mj {, and let Y it = tanh" 1 r tf, where r it is the sample correlation coefficient for the jth subsample of the tth sample; for our Monte-Carlo simulation, m = 5. Under H^, the T ti are independently distributed with equal means and variances. The procedure is to perform a two-sample t test based on the Y {j. The robustness of the (test suggests that this method should perform well for small samples with respect to significance level. Under Hg, the test statistic is asymptotically standard normal, irrespective of the distributions of the populations. The second asymptotically robust test we examine, designated the jackknife test, is based on the jackknife technique; see Miller (1964) and the references given there. Let r^ be the correlation coefficient calculated from the tth sample omitting the jth observation. Let U i} = n i t&dhr 1 r i (rif ljtanh" 1^. We treat the U i} as being approximately independent and having, under H o> approximately equal means and variances, and we test H^ using a two-sample (test. The results of the Monte-Carlo experiments in 2 show that this method works reasonably well for small samples when the sample sizes are equal. For equal sample sizes, the means and variances of the Uy are exactly equal under HQ. We have not investigated the case of unequal sample sizes in the present Monte-Carlo study, but we would Downloaded from at Penn State University (Paterno Lib) on May 9, 216

3 Asymptotically robust tests for correlation coefficients 553 not anticipate a problem from this source for reasonably large samples. It can be shown (Arvesen, 1969), that under HQ the asymptotic distribution of the test statistic is standard normal, for any population distribution. 2. MONTE-CABLO SIMULATION Small sample results for the tests described in 1 were obtained by simulating on a Burroughs 67 computer random samples from six bivariate distributions with various kurtoses; see Table 1. A mixed congruential pseudo-random number generator was used to produce independent random variables U t, distributed uniformly on the interval (,1). Standard normal random variables Z i were simulated by JJ^ U lt 6. Bivariate random vectors (X, Y) were then constructed to have population correlation coefficient p in the following manner: 1. Normal: X = Z lt Y = pzj^ + Z^l-p 2 ); 2. Log normal: X = exp(z 1 ), Y= expft^ + Zj^l-T 2 )}, where T = log{/o(e- 1) + 1}; 3. Linear regression uniform: X = U v Y = p(l p 2 )-iu 1 +U i ; the marginal distribution of Y is not uniform and the regression of Y on X is linear; 4. Linear regression double exponential: X = ±logu v each with probability, Y = p(l p*)-l X + e, where e = + log U 2, each with probability ; the marginal distribution of Y is not double exponential, and the regression of 7 on 1 is linear; 5. Gamma: marginals ^2 1 1(i x = -x\ogu t, r = - ] This is the bivariate gamma distribution, due to Cherian (1941), which is studied by David & Fix (1961); it has linear regressions. 6. Contaminated normal: ( fz 1,pZ 1 + Z 2 J(l-p i ) (with probability -9), ' ' ~ \ic{z 1,pZ 1 + Z 2 J(l-p t )} (with probability -1). Generally replications of the experiment of sampling 25 observations from each of two bivariate populations of the same distributional family were made. David (1938) suggests 25 observations as a lower bound for the adequacy of the normal approximation to the distribution of Fisher's transformation of the sample correlation coefficient, z = tanh -1 r, under normality. These simulated samples were used to examine the small-sample behaviour of the procedures described in 1 both for testing equality of correlation coefficients and for construction of confidence intervals for a single correlation coefficient. Asymptotic theory suggests that for some of the distributions sampled, one might anticipate substantial departures from the nominal significance levels derived under normal theory for the test based on Fisher's z. The first column of Table 1 gives the numerical values of the kurtosis of each of the marginal distributions sampled in the Monte-Carlo study. The kurtosis is quite large for the log normal distribution and negative for the linear regression uniform distribution. The asymptotic variance of ^/»tanh~ 1 r is given numerically in the second column of Table 1 and demonstrates considerable deviation from the normal theory nominal value of 1. General expressions for the asymptotic variance are given below. The last two columns of Table 1 give the asymptotic size, valid for both one sample and two sample normal theory tests, for the nominal 5 % level. Downloaded from at Penn State University (Paterno Lib) on May 9, 216

4 554 G. T. DUNCAN AND M. W. J. LAYABD Table 1. test. Asymptotic results for distributions in sampling study Distribution p = -6 p = -9 jfc = 3 p = 6 p = -9 p = -6 p = -9 p = -6 p = -9 Kurto8is of marginal distributions* Log normal Asymptotic variance of Jn tanh" 1 r Contaminated normal Linear regression uniform -1-2 (--924) -1-2 (-1-15) Linear regression double 3 (2-31) 3- (2-537) -6-6 Gamma exponential Asymptotic size; one-sample and two-sample tests, nominal 6 % level * Kurtosis for 'linear regression' distributions ia for X component; kurtosis for Y component is given in parentheses. Let o- 2 (p)/n denote the asymptotic variance of z as a function of the true correlation p. Then for the contaminated normal distribution, **( P ) = For each of the linear regression distributions with Y = p(l ^!)~H + e, o" 2 (p) = where X and e each have kurtosis A. For the log normal distribution, a 2 (p) is given by (l-l), with A 31 = A 13 = 7 5- (e 1... _ i2e-6), A M = (e 1) Lastly, for the bivariate gamma distribution, 1-2/(2-/) P - e 3 + 6e - 6), - e 2 + 4e - 6). Downloaded from at Penn State University (Paterno Lib) on May 9, 216 Note that for each of the bivariate distributions sampled, except the contaminated normal, p = implies independence, and for these distributions the asymptotic variance of z is 1/n when p =. But for the contaminated normal the components are not independent when p =, and the asymptotic variance of z is not 1/n in that case; nor does the asymptotic variance depend on p.

5 Asymptotically robust tests for correlation coefficients 555 Tables 2 and 3 give the Monte-Carlo results for the (m = 5) and jackknife procedures, and for the normal theory procedures based on Fisher's z. We first discuss the one-sample case and then the two-sample case. Table 2 gives simulation results for one sample confidence interval procedures. Associated with each of the test procedures is a method for constructing confidence intervals for a single correlation coefficient. Each entry in the first three columns of Table 2 is the observed fraction of coverages of the true correlation coefficient. We note that subtracting this fraction from 1 gives the empirical size of a two-sided test for equality with the stated correlation coefficient. Also note that for 1 replications the standard deviation of the fraction of coverages is approximately -6, when the true probability of coverage is about -95. Table 2. Empirical results for nominal 95% confidence intervals for a single correlation coefficient, p. For the test, m = 5 Distribution sampled Normal, n = 26 Log normal, n = 26 Linear regression uniform, n = 26 Linear regression double exponential, n = 25 Gamma, n = 25 Contaminated normal, k = 3, n = 1 Contaminated normal, k = 3, n = 25 Contaminated normal, k = 3, n = 1 Correlation coefficient Replications Fraction of coverages of p Downloaded from at Penn State University (Paterno Lib) on May 9, 216 For tests of equality of correlation coefficients, bias of the test statistics is not a problem since under the null hypothesis these test statistics are simply differences of two random variables with the same distribution, and hence have mean. But in constructing a confidence interval for a single correlation coefficient, bias of the relevant statistic may be a problem, especially for the procedure, which in our simulation studies is based on subsamples of only 5 observations. For the bivariate normal distribution, the bias of tanh -1 r is approximately \p{n I)" 1 ; bias of this order is removed by the jackknife.

6 556 G. T. DUNCAN AND M. W. J. LAYABD We now discuss the results in Table 2, which give the empirical coverage of nominal 95 % confidence intervals for p. Although not shown, the empirical average length was also calculated and will be discussed. (1). The normal theory procedure maintains the nominal confidence level -95 for every value of p only under normality and in the Linear regression uniform case. However, its performance is generally adequate for p =, except for the contaminated normal distribution. (2). The jacfcknife does reasonably well in maintaining nominal levels in the uniform, double exponential, gamma and contaminated normal cases, but rather poorly in the log normal case. For the normal distributions sampled, the average length of the jackknife intervals is slightly more than that of the normal theory intervals. (3). The effects of bias are clearly apparent for nonzero correlations. For p =, the intervals have about the right confidence level, but the average lengths are in most cases considerably greater than those of the jackknife intervals. We conclude that the procedure is unsuitable for constructing confidence intervals for a single correlation coefficient, or for testing hypotheses about a single coefficient. In Table 3, empirical test sizes at the nominal 5 % level are given for when {p lt p 2 ) = (,), (-6, -6) and (-9, -9). Empirical power is given for (p v p 2 ) = (,-6) and (,-9). In the contaminated normal case sample sizes of 1 and 1 were also examined; for the sample size of 1 power calculations were made for (p^pt) = (,-3) and (,-6). Since replications were made, the standard deviation of any entry in Table 3 is -1 for true probability of rejection -5, and it is never higher than 22, which is the standard deviation for true probability of rejection -5. Results at the nominal 1 % level were computed but are not shown; they are consistent with those in Table 3. The results in Table 3 are now discussed for each distribution. (1) Normal. The size of each of the 3 tests does not deviate significantly from nominal levels. The power of the test lags, but the other tests have comparable power. (2) Log normal. For this skew distribution with a heavy tail, the normal theory test fails badly with actual sizes differing markedly from nominal, particularly as the common correlation increases. The test works well in maintaining the nominal level and its power is comparable to that of the jaokknife. The significance levels of the jackknife test are higher than nominal for nonzero correlations, but are much lower than those of the normaltheory test. (3) Linear regression uniform. The significance level of the normal-theory test is below nominal for nonzero correlation, but its power is as good as the jackknife test. The test has relatively poor power. (4) Linear regression double exponential. The size of the normal-theory test departs increasingly from nominal levels as the common correlation gets larger. The power of the jackknife test is a little better than that of the test. (5) Oamma. The significance level of the normal-theory test is too high for a large common correlation. The jackknife has better power than. (6) Contaminated normal. The normal-theory test continues to fare badly, and it is interesting to note that it does worse as the sample size increases. Both the test and Downloaded from at Penn State University (Paterno Lib) on May 9, 216

7 Asymptotically robust tests for correlation coefficients 557 Table 3. Empirical size and power based on replications for tests of equality of two correlation coefficients at nominal 5 % level. For the test, m = 5 (Pi' Pt) (,) (-6,-6) (-9, -9) Normal; n = Log normal; n = Linear regression uniform ;n = Linear regression double exponential; n = Gamma; n = Contaminated normal; k = Contaminated normal; k = Contaminated normal; k = , n = , n = , n = (, -6) (, -9) the jackknife test have about the right size but the jackknife is slightly better on grounds of power. The results of Table 3 suggest the following general conclusions: (i) the normal theory test can be strongly affected by nonnormality, even if the correlations are small; (ii) the test performs very well in maintaining nominal levels but may have low power; (iii) the jackknife does reasonably well in maintaining nominal size and its power is not much less than the best of the three tests for each bivariate distribution sampled. Downloaded from at Penn State University (Paterno Lib) on May 9, 216

8 558 G. T. DUNCAN AND M. W. J. LAYARD 3. CONCLUSIONS Consistent with asymptotic results, the normal theory procedure based on Fisher's z has been shown to produce poor results for certain nonnormal distributions. The test does well in maintaining nominal levels for tests of equality of two correlation coemcients, but often has low power. The small sample bias of the test makes it unsuitable for tests and confidence intervals about a single correlation coefficient. The jackknife procedure applied to Fisher's z gives a test which comes much closer than the normal theory method to maintaining nominal levels while having comparable power. It should be noted that our sampling results are consistent with the implications of Pitman's (1937) study in the limited case where zero correlation implies independence. Specifically, Pitman shows that under independence the normal theory test is a good approximation to a permutation test and hence would be expected to be robust. As we remarked in 1, the asymptotic variance is correct in this special case and the sampling results for normal theory tests for all distributions meeting this independence condition are close to nominal. However, the condition is not met for the contaminated normal distribution and the empirical results are far from nominal. REFERENCES ABVESEN, J. N. (1969). Jackknifing {/-statistics. Ann. Math. Statist. 4, BAXHB, G. A. (193). The significance of the product-moment coefficient of correlation with special reference to the character of the marginal distributions. J. Am. Statist. Ass. 25, , G. E. P. (1953). Non-normality and tests on variances. Biometrika 4, CHEBIAN, K. C. (1941). A bivariate correlated gamma-type distribution function. J. Ind. Math. Soc. 5, DAVID, F. N. (1938). Tables of the Correlation Coefficient. Cambridge University Press. DAVID, F. N. & Fix, E. (1961). Rank correlation and regression in a non-normal surface. Proc. 4th Berkeley Symp. 1, KOWALSKI, C. J. (1972). On the effecte of non-normality on the distribution of the sample productmoment correlation coefficient. Appl. Statist. 21, MTTT.TCB, FV. G. (1964). A trustworthy jackknife. Ann. Math. Statist. 35, NEFZGEB, M. D. & DBASQOW, J. (1957). The needless assumption of normality in Pearson's r. Am. Psychol. 12, NOBBIS, R. C. & HEIJM, H. F. (1961). Non-normality and product-moment correlation. J. Exper. Educ. 29, PEARSON, E. S. (1931). The test of significance for the correlation coefficient. J. Am. Statist. Ass. 26, PEABSON, E. S. (1932). The test of significance for the correlation coefficient: some further results. J. Am. Statist. Ass. 27, PITMAN, E. J. G. (1937). Significance tests which may be applied to samples from any populations. n. The correlation coefficient test. J. R. Statist. Son., Suppl. 4, pp Downloaded from at Penn State University (Paterno Lib) on May 9, 216

TESTING FOR NORMALITY IN THE LINEAR REGRESSION MODEL: AN EMPIRICAL LIKELIHOOD RATIO TEST

TESTING FOR NORMALITY IN THE LINEAR REGRESSION MODEL: AN EMPIRICAL LIKELIHOOD RATIO TEST Econometrics Working Paper EWP0402 ISSN 1485-6441 Department of Economics TESTING FOR NORMALITY IN THE LINEAR REGRESSION MODEL: AN EMPIRICAL LIKELIHOOD RATIO TEST Lauren Bin Dong & David E. A. Giles Department

More information

Research Article A Nonparametric Two-Sample Wald Test of Equality of Variances

Research Article A Nonparametric Two-Sample Wald Test of Equality of Variances Advances in Decision Sciences Volume 211, Article ID 74858, 8 pages doi:1.1155/211/74858 Research Article A Nonparametric Two-Sample Wald Test of Equality of Variances David Allingham 1 andj.c.w.rayner

More information

MA 575 Linear Models: Cedric E. Ginestet, Boston University Non-parametric Inference, Polynomial Regression Week 9, Lecture 2

MA 575 Linear Models: Cedric E. Ginestet, Boston University Non-parametric Inference, Polynomial Regression Week 9, Lecture 2 MA 575 Linear Models: Cedric E. Ginestet, Boston University Non-parametric Inference, Polynomial Regression Week 9, Lecture 2 1 Bootstrapped Bias and CIs Given a multiple regression model with mean and

More information

A nonparametric two-sample wald test of equality of variances

A nonparametric two-sample wald test of equality of variances University of Wollongong Research Online Faculty of Informatics - Papers (Archive) Faculty of Engineering and Information Sciences 211 A nonparametric two-sample wald test of equality of variances David

More information

Application of Parametric Homogeneity of Variances Tests under Violation of Classical Assumption

Application of Parametric Homogeneity of Variances Tests under Violation of Classical Assumption Application of Parametric Homogeneity of Variances Tests under Violation of Classical Assumption Alisa A. Gorbunova and Boris Yu. Lemeshko Novosibirsk State Technical University Department of Applied Mathematics,

More information

NAG Library Chapter Introduction. G08 Nonparametric Statistics

NAG Library Chapter Introduction. G08 Nonparametric Statistics NAG Library Chapter Introduction G08 Nonparametric Statistics Contents 1 Scope of the Chapter.... 2 2 Background to the Problems... 2 2.1 Parametric and Nonparametric Hypothesis Testing... 2 2.2 Types

More information

The Nonparametric Bootstrap

The Nonparametric Bootstrap The Nonparametric Bootstrap The nonparametric bootstrap may involve inferences about a parameter, but we use a nonparametric procedure in approximating the parametric distribution using the ECDF. We use

More information

Psychology 282 Lecture #4 Outline Inferences in SLR

Psychology 282 Lecture #4 Outline Inferences in SLR Psychology 282 Lecture #4 Outline Inferences in SLR Assumptions To this point we have not had to make any distributional assumptions. Principle of least squares requires no assumptions. Can use correlations

More information

APPLICATION AND POWER OF PARAMETRIC CRITERIA FOR TESTING THE HOMOGENEITY OF VARIANCES. PART IV

APPLICATION AND POWER OF PARAMETRIC CRITERIA FOR TESTING THE HOMOGENEITY OF VARIANCES. PART IV DOI 10.1007/s11018-017-1213-4 Measurement Techniques, Vol. 60, No. 5, August, 2017 APPLICATION AND POWER OF PARAMETRIC CRITERIA FOR TESTING THE HOMOGENEITY OF VARIANCES. PART IV B. Yu. Lemeshko and T.

More information

Analysis of variance and linear contrasts in experimental design with generalized secant hyperbolic distribution

Analysis of variance and linear contrasts in experimental design with generalized secant hyperbolic distribution Journal of Computational and Applied Mathematics 216 (2008) 545 553 www.elsevier.com/locate/cam Analysis of variance and linear contrasts in experimental design with generalized secant hyperbolic distribution

More information

CHAPTER 17 CHI-SQUARE AND OTHER NONPARAMETRIC TESTS FROM: PAGANO, R. R. (2007)

CHAPTER 17 CHI-SQUARE AND OTHER NONPARAMETRIC TESTS FROM: PAGANO, R. R. (2007) FROM: PAGANO, R. R. (007) I. INTRODUCTION: DISTINCTION BETWEEN PARAMETRIC AND NON-PARAMETRIC TESTS Statistical inference tests are often classified as to whether they are parametric or nonparametric Parameter

More information

401 Review. 6. Power analysis for one/two-sample hypothesis tests and for correlation analysis.

401 Review. 6. Power analysis for one/two-sample hypothesis tests and for correlation analysis. 401 Review Major topics of the course 1. Univariate analysis 2. Bivariate analysis 3. Simple linear regression 4. Linear algebra 5. Multiple regression analysis Major analysis methods 1. Graphical analysis

More information

Questions 3.83, 6.11, 6.12, 6.17, 6.25, 6.29, 6.33, 6.35, 6.50, 6.51, 6.53, 6.55, 6.59, 6.60, 6.65, 6.69, 6.70, 6.77, 6.79, 6.89, 6.

Questions 3.83, 6.11, 6.12, 6.17, 6.25, 6.29, 6.33, 6.35, 6.50, 6.51, 6.53, 6.55, 6.59, 6.60, 6.65, 6.69, 6.70, 6.77, 6.79, 6.89, 6. Chapter 7 Reading 7.1, 7.2 Questions 3.83, 6.11, 6.12, 6.17, 6.25, 6.29, 6.33, 6.35, 6.50, 6.51, 6.53, 6.55, 6.59, 6.60, 6.65, 6.69, 6.70, 6.77, 6.79, 6.89, 6.112 Introduction In Chapter 5 and 6, we emphasized

More information

A correlation coefficient for circular data

A correlation coefficient for circular data BiomelriL-a (1983). 70. 2, pp. 327-32 327 Prinltd in Great Britain A correlation coefficient for circular data BY N. I. FISHER CSIRO Division of Mathematics and Statistics, Lindfield, N.S.W., Australia

More information

Multiple Comparison Procedures, Trimmed Means and Transformed Statistics. Rhonda K. Kowalchuk Southern Illinois University Carbondale

Multiple Comparison Procedures, Trimmed Means and Transformed Statistics. Rhonda K. Kowalchuk Southern Illinois University Carbondale Multiple Comparison Procedures 1 Multiple Comparison Procedures, Trimmed Means and Transformed Statistics Rhonda K. Kowalchuk Southern Illinois University Carbondale H. J. Keselman University of Manitoba

More information

CONVERTING OBSERVED LIKELIHOOD FUNCTIONS TO TAIL PROBABILITIES. D.A.S. Fraser Mathematics Department York University North York, Ontario M3J 1P3

CONVERTING OBSERVED LIKELIHOOD FUNCTIONS TO TAIL PROBABILITIES. D.A.S. Fraser Mathematics Department York University North York, Ontario M3J 1P3 CONVERTING OBSERVED LIKELIHOOD FUNCTIONS TO TAIL PROBABILITIES D.A.S. Fraser Mathematics Department York University North York, Ontario M3J 1P3 N. Reid Department of Statistics University of Toronto Toronto,

More information

11. Bootstrap Methods

11. Bootstrap Methods 11. Bootstrap Methods c A. Colin Cameron & Pravin K. Trivedi 2006 These transparencies were prepared in 20043. They can be used as an adjunct to Chapter 11 of our subsequent book Microeconometrics: Methods

More information

Two-by-two ANOVA: Global and Graphical Comparisons Based on an Extension of the Shift Function

Two-by-two ANOVA: Global and Graphical Comparisons Based on an Extension of the Shift Function Journal of Data Science 7(2009), 459-468 Two-by-two ANOVA: Global and Graphical Comparisons Based on an Extension of the Shift Function Rand R. Wilcox University of Southern California Abstract: When comparing

More information

ON LARGE SAMPLE PROPERTIES OF CERTAIN NONPARAMETRIC PROCEDURES

ON LARGE SAMPLE PROPERTIES OF CERTAIN NONPARAMETRIC PROCEDURES ON LARGE SAMPLE PROPERTIES OF CERTAIN NONPARAMETRIC PROCEDURES 1. Summary and introduction HERMAN RUBIN PURDUE UNIVERSITY Efficiencies of one sided and two sided procedures are considered from the standpoint

More information

Supporting Information for Estimating restricted mean. treatment effects with stacked survival models

Supporting Information for Estimating restricted mean. treatment effects with stacked survival models Supporting Information for Estimating restricted mean treatment effects with stacked survival models Andrew Wey, David Vock, John Connett, and Kyle Rudser Section 1 presents several extensions to the simulation

More information

Application of Variance Homogeneity Tests Under Violation of Normality Assumption

Application of Variance Homogeneity Tests Under Violation of Normality Assumption Application of Variance Homogeneity Tests Under Violation of Normality Assumption Alisa A. Gorbunova, Boris Yu. Lemeshko Novosibirsk State Technical University Novosibirsk, Russia e-mail: gorbunova.alisa@gmail.com

More information

A Monte Carlo Simulation of the Robust Rank- Order Test Under Various Population Symmetry Conditions

A Monte Carlo Simulation of the Robust Rank- Order Test Under Various Population Symmetry Conditions Journal of Modern Applied Statistical Methods Volume 12 Issue 1 Article 7 5-1-2013 A Monte Carlo Simulation of the Robust Rank- Order Test Under Various Population Symmetry Conditions William T. Mickelson

More information

AN EMPIRICAL LIKELIHOOD RATIO TEST FOR NORMALITY

AN EMPIRICAL LIKELIHOOD RATIO TEST FOR NORMALITY Econometrics Working Paper EWP0401 ISSN 1485-6441 Department of Economics AN EMPIRICAL LIKELIHOOD RATIO TEST FOR NORMALITY Lauren Bin Dong & David E. A. Giles Department of Economics, University of Victoria

More information

Estimation of the Conditional Variance in Paired Experiments

Estimation of the Conditional Variance in Paired Experiments Estimation of the Conditional Variance in Paired Experiments Alberto Abadie & Guido W. Imbens Harvard University and BER June 008 Abstract In paired randomized experiments units are grouped in pairs, often

More information

HANDBOOK OF APPLICABLE MATHEMATICS

HANDBOOK OF APPLICABLE MATHEMATICS HANDBOOK OF APPLICABLE MATHEMATICS Chief Editor: Walter Ledermann Volume VI: Statistics PART A Edited by Emlyn Lloyd University of Lancaster A Wiley-Interscience Publication JOHN WILEY & SONS Chichester

More information

Extending the Robust Means Modeling Framework. Alyssa Counsell, Phil Chalmers, Matt Sigal, Rob Cribbie

Extending the Robust Means Modeling Framework. Alyssa Counsell, Phil Chalmers, Matt Sigal, Rob Cribbie Extending the Robust Means Modeling Framework Alyssa Counsell, Phil Chalmers, Matt Sigal, Rob Cribbie One-way Independent Subjects Design Model: Y ij = µ + τ j + ε ij, j = 1,, J Y ij = score of the ith

More information

STAT Section 5.8: Block Designs

STAT Section 5.8: Block Designs STAT 518 --- Section 5.8: Block Designs Recall that in paired-data studies, we match up pairs of subjects so that the two subjects in a pair are alike in some sense. Then we randomly assign, say, treatment

More information

Comparison of Power between Adaptive Tests and Other Tests in the Field of Two Sample Scale Problem

Comparison of Power between Adaptive Tests and Other Tests in the Field of Two Sample Scale Problem Comparison of Power between Adaptive Tests and Other Tests in the Field of Two Sample Scale Problem Chikhla Jun Gogoi 1, Dr. Bipin Gogoi 2 1 Research Scholar, Department of Statistics, Dibrugarh University,

More information

Fall 2017 STAT 532 Homework Peter Hoff. 1. Let P be a probability measure on a collection of sets A.

Fall 2017 STAT 532 Homework Peter Hoff. 1. Let P be a probability measure on a collection of sets A. 1. Let P be a probability measure on a collection of sets A. (a) For each n N, let H n be a set in A such that H n H n+1. Show that P (H n ) monotonically converges to P ( k=1 H k) as n. (b) For each n

More information

Joint Estimation of Risk Preferences and Technology: Further Discussion

Joint Estimation of Risk Preferences and Technology: Further Discussion Joint Estimation of Risk Preferences and Technology: Further Discussion Feng Wu Research Associate Gulf Coast Research and Education Center University of Florida Zhengfei Guan Assistant Professor Gulf

More information

A better way to bootstrap pairs

A better way to bootstrap pairs A better way to bootstrap pairs Emmanuel Flachaire GREQAM - Université de la Méditerranée CORE - Université Catholique de Louvain April 999 Abstract In this paper we are interested in heteroskedastic regression

More information

B.N.Bandodkar College of Science, Thane. Random-Number Generation. Mrs M.J.Gholba

B.N.Bandodkar College of Science, Thane. Random-Number Generation. Mrs M.J.Gholba B.N.Bandodkar College of Science, Thane Random-Number Generation Mrs M.J.Gholba Properties of Random Numbers A sequence of random numbers, R, R,., must have two important statistical properties, uniformity

More information

Biometrika Trust. Biometrika Trust is collaborating with JSTOR to digitize, preserve and extend access to Biometrika.

Biometrika Trust. Biometrika Trust is collaborating with JSTOR to digitize, preserve and extend access to Biometrika. Biometrika Trust An Improved Bonferroni Procedure for Multiple Tests of Significance Author(s): R. J. Simes Source: Biometrika, Vol. 73, No. 3 (Dec., 1986), pp. 751-754 Published by: Biometrika Trust Stable

More information

Least Absolute Value vs. Least Squares Estimation and Inference Procedures in Regression Models with Asymmetric Error Distributions

Least Absolute Value vs. Least Squares Estimation and Inference Procedures in Regression Models with Asymmetric Error Distributions Journal of Modern Applied Statistical Methods Volume 8 Issue 1 Article 13 5-1-2009 Least Absolute Value vs. Least Squares Estimation and Inference Procedures in Regression Models with Asymmetric Error

More information

Confidence Interval Estimation

Confidence Interval Estimation Department of Psychology and Human Development Vanderbilt University 1 Introduction 2 3 4 5 Relationship to the 2-Tailed Hypothesis Test Relationship to the 1-Tailed Hypothesis Test 6 7 Introduction In

More information

STATISTICS ANCILLARY SYLLABUS. (W.E.F. the session ) Semester Paper Code Marks Credits Topic

STATISTICS ANCILLARY SYLLABUS. (W.E.F. the session ) Semester Paper Code Marks Credits Topic STATISTICS ANCILLARY SYLLABUS (W.E.F. the session 2014-15) Semester Paper Code Marks Credits Topic 1 ST21012T 70 4 Descriptive Statistics 1 & Probability Theory 1 ST21012P 30 1 Practical- Using Minitab

More information

Inferences on a Normal Covariance Matrix and Generalized Variance with Monotone Missing Data

Inferences on a Normal Covariance Matrix and Generalized Variance with Monotone Missing Data Journal of Multivariate Analysis 78, 6282 (2001) doi:10.1006jmva.2000.1939, available online at http:www.idealibrary.com on Inferences on a Normal Covariance Matrix and Generalized Variance with Monotone

More information

A TIME SERIES PARADOX: UNIT ROOT TESTS PERFORM POORLY WHEN DATA ARE COINTEGRATED

A TIME SERIES PARADOX: UNIT ROOT TESTS PERFORM POORLY WHEN DATA ARE COINTEGRATED A TIME SERIES PARADOX: UNIT ROOT TESTS PERFORM POORLY WHEN DATA ARE COINTEGRATED by W. Robert Reed Department of Economics and Finance University of Canterbury, New Zealand Email: bob.reed@canterbury.ac.nz

More information

The exact bootstrap method shown on the example of the mean and variance estimation

The exact bootstrap method shown on the example of the mean and variance estimation Comput Stat (2013) 28:1061 1077 DOI 10.1007/s00180-012-0350-0 ORIGINAL PAPER The exact bootstrap method shown on the example of the mean and variance estimation Joanna Kisielinska Received: 21 May 2011

More information

STATISTICS SYLLABUS UNIT I

STATISTICS SYLLABUS UNIT I STATISTICS SYLLABUS UNIT I (Probability Theory) Definition Classical and axiomatic approaches.laws of total and compound probability, conditional probability, Bayes Theorem. Random variable and its distribution

More information

Conventional And Robust Paired And Independent-Samples t Tests: Type I Error And Power Rates

Conventional And Robust Paired And Independent-Samples t Tests: Type I Error And Power Rates Journal of Modern Applied Statistical Methods Volume Issue Article --3 Conventional And And Independent-Samples t Tests: Type I Error And Power Rates Katherine Fradette University of Manitoba, umfradet@cc.umanitoba.ca

More information

BIOL 4605/7220 CH 20.1 Correlation

BIOL 4605/7220 CH 20.1 Correlation BIOL 4605/70 CH 0. Correlation GPT Lectures Cailin Xu November 9, 0 GLM: correlation Regression ANOVA Only one dependent variable GLM ANCOVA Multivariate analysis Multiple dependent variables (Correlation)

More information

AP Statistics Cumulative AP Exam Study Guide

AP Statistics Cumulative AP Exam Study Guide AP Statistics Cumulative AP Eam Study Guide Chapters & 3 - Graphs Statistics the science of collecting, analyzing, and drawing conclusions from data. Descriptive methods of organizing and summarizing statistics

More information

ON THE CONSEQUENCES OF MISSPECIFING ASSUMPTIONS CONCERNING RESIDUALS DISTRIBUTION IN A REPEATED MEASURES AND NONLINEAR MIXED MODELLING CONTEXT

ON THE CONSEQUENCES OF MISSPECIFING ASSUMPTIONS CONCERNING RESIDUALS DISTRIBUTION IN A REPEATED MEASURES AND NONLINEAR MIXED MODELLING CONTEXT ON THE CONSEQUENCES OF MISSPECIFING ASSUMPTIONS CONCERNING RESIDUALS DISTRIBUTION IN A REPEATED MEASURES AND NONLINEAR MIXED MODELLING CONTEXT Rachid el Halimi and Jordi Ocaña Departament d Estadística

More information

Dr. Junchao Xia Center of Biophysics and Computational Biology. Fall /1/2016 1/46

Dr. Junchao Xia Center of Biophysics and Computational Biology. Fall /1/2016 1/46 BIO5312 Biostatistics Lecture 10:Regression and Correlation Methods Dr. Junchao Xia Center of Biophysics and Computational Biology Fall 2016 11/1/2016 1/46 Outline In this lecture, we will discuss topics

More information

Chapter 1 Statistical Inference

Chapter 1 Statistical Inference Chapter 1 Statistical Inference causal inference To infer causality, you need a randomized experiment (or a huge observational study and lots of outside information). inference to populations Generalizations

More information

INFLUENCE OF USING ALTERNATIVE MEANS ON TYPE-I ERROR RATE IN THE COMPARISON OF INDEPENDENT GROUPS ABSTRACT

INFLUENCE OF USING ALTERNATIVE MEANS ON TYPE-I ERROR RATE IN THE COMPARISON OF INDEPENDENT GROUPS ABSTRACT Mirtagioğlu et al., The Journal of Animal & Plant Sciences, 4(): 04, Page: J. 344-349 Anim. Plant Sci. 4():04 ISSN: 08-708 INFLUENCE OF USING ALTERNATIVE MEANS ON TYPE-I ERROR RATE IN THE COMPARISON OF

More information

Fast and robust bootstrap for LTS

Fast and robust bootstrap for LTS Fast and robust bootstrap for LTS Gert Willems a,, Stefan Van Aelst b a Department of Mathematics and Computer Science, University of Antwerp, Middelheimlaan 1, B-2020 Antwerp, Belgium b Department of

More information

Subject CS1 Actuarial Statistics 1 Core Principles

Subject CS1 Actuarial Statistics 1 Core Principles Institute of Actuaries of India Subject CS1 Actuarial Statistics 1 Core Principles For 2019 Examinations Aim The aim of the Actuarial Statistics 1 subject is to provide a grounding in mathematical and

More information

THE MAYFIELD METHOD OF ESTIMATING NESTING SUCCESS: A MODEL, ESTIMATORS AND SIMULATION RESULTS

THE MAYFIELD METHOD OF ESTIMATING NESTING SUCCESS: A MODEL, ESTIMATORS AND SIMULATION RESULTS Wilson. Bull., 93(l), 1981, pp. 42-53 THE MAYFIELD METHOD OF ESTIMATING NESTING SUCCESS: A MODEL, ESTIMATORS AND SIMULATION RESULTS GARY L. HENSLER AND JAMES D. NICHOLS Mayfield (1960, 1961, 1975) proposed

More information

Better Bootstrap Confidence Intervals

Better Bootstrap Confidence Intervals by Bradley Efron University of Washington, Department of Statistics April 12, 2012 An example Suppose we wish to make inference on some parameter θ T (F ) (e.g. θ = E F X ), based on data We might suppose

More information

Efficient Robbins-Monro Procedure for Binary Data

Efficient Robbins-Monro Procedure for Binary Data Efficient Robbins-Monro Procedure for Binary Data V. Roshan Joseph School of Industrial and Systems Engineering Georgia Institute of Technology Atlanta, GA 30332-0205, USA roshan@isye.gatech.edu SUMMARY

More information

Comparing Two Dependent Groups: Dealing with Missing Values

Comparing Two Dependent Groups: Dealing with Missing Values Journal of Data Science 9(2011), 1-13 Comparing Two Dependent Groups: Dealing with Missing Values Rand R. Wilcox University of Southern California Abstract: The paper considers the problem of comparing

More information

Bootstrapping Heteroskedasticity Consistent Covariance Matrix Estimator

Bootstrapping Heteroskedasticity Consistent Covariance Matrix Estimator Bootstrapping Heteroskedasticity Consistent Covariance Matrix Estimator by Emmanuel Flachaire Eurequa, University Paris I Panthéon-Sorbonne December 2001 Abstract Recent results of Cribari-Neto and Zarkos

More information

Methodology Review: Applications of Distribution Theory in Studies of. Population Validity and Cross Validity. James Algina. University of Florida

Methodology Review: Applications of Distribution Theory in Studies of. Population Validity and Cross Validity. James Algina. University of Florida Distribution Theory 1 Methodology eview: Applications of Distribution Theory in Studies of Population Validity and Cross Validity by James Algina University of Florida and H. J. Keselman University of

More information

Journal of Biostatistics and Epidemiology

Journal of Biostatistics and Epidemiology Journal of Biostatistics and Epidemiology Original Article Robust correlation coefficient goodness-of-fit test for the Gumbel distribution Abbas Mahdavi 1* 1 Department of Statistics, School of Mathematical

More information

Journal of Educational and Behavioral Statistics

Journal of Educational and Behavioral Statistics Journal of Educational and Behavioral Statistics http://jebs.aera.net Theory of Estimation and Testing of Effect Sizes: Use in Meta-Analysis Helena Chmura Kraemer JOURNAL OF EDUCATIONAL AND BEHAVIORAL

More information

YUN WU. B.S., Gui Zhou University of Finance and Economics, 1996 A REPORT. Submitted in partial fulfillment of the requirements for the degree

YUN WU. B.S., Gui Zhou University of Finance and Economics, 1996 A REPORT. Submitted in partial fulfillment of the requirements for the degree A SIMULATION STUDY OF THE ROBUSTNESS OF HOTELLING S T TEST FOR THE MEAN OF A MULTIVARIATE DISTRIBUTION WHEN SAMPLING FROM A MULTIVARIATE SKEW-NORMAL DISTRIBUTION by YUN WU B.S., Gui Zhou University of

More information

Institute of Actuaries of India

Institute of Actuaries of India Institute of Actuaries of India Subject CT3 Probability and Mathematical Statistics For 2018 Examinations Subject CT3 Probability and Mathematical Statistics Core Technical Syllabus 1 June 2017 Aim The

More information

NAG Library Chapter Introduction. g08 Nonparametric Statistics

NAG Library Chapter Introduction. g08 Nonparametric Statistics g08 Nonparametric Statistics Introduction g08 NAG Library Chapter Introduction g08 Nonparametric Statistics Contents 1 Scope of the Chapter.... 2 2 Background to the Problems... 2 2.1 Parametric and Nonparametric

More information

A New Procedure for Multiple Testing of Econometric Models

A New Procedure for Multiple Testing of Econometric Models A New Procedure for Multiple Testing of Econometric Models Maxwell L. King 1, Xibin Zhang, and Muhammad Akram Department of Econometrics and Business Statistics Monash University, Australia April 2007

More information

An Overview of the Performance of Four Alternatives to Hotelling's T Square

An Overview of the Performance of Four Alternatives to Hotelling's T Square fi~hjf~~ G 1992, m-t~, 11o-114 Educational Research Journal 1992, Vol.7, pp. 110-114 An Overview of the Performance of Four Alternatives to Hotelling's T Square LIN Wen-ying The Chinese University of Hong

More information

Model Selection, Estimation, and Bootstrap Smoothing. Bradley Efron Stanford University

Model Selection, Estimation, and Bootstrap Smoothing. Bradley Efron Stanford University Model Selection, Estimation, and Bootstrap Smoothing Bradley Efron Stanford University Estimation After Model Selection Usually: (a) look at data (b) choose model (linear, quad, cubic...?) (c) fit estimates

More information

6 Single Sample Methods for a Location Parameter

6 Single Sample Methods for a Location Parameter 6 Single Sample Methods for a Location Parameter If there are serious departures from parametric test assumptions (e.g., normality or symmetry), nonparametric tests on a measure of central tendency (usually

More information

On Polynomial Transformations For Simulating Multivariate Non-normal Distributions

On Polynomial Transformations For Simulating Multivariate Non-normal Distributions Journal of Modern Applied Statistical Methods May, 2004, Vol. 3, No. 1,65-71 Copyright 2004 JMASM, Inc. 1538-9472/04/$95.00 On Polynomial Transformations For Simulating Multivariate Non-normal Distributions

More information

Semester , Example Exam 1

Semester , Example Exam 1 Semester 1 2017, Example Exam 1 1 of 10 Instructions The exam consists of 4 questions, 1-4. Each question has four items, a-d. Within each question: Item (a) carries a weight of 8 marks. Item (b) carries

More information

Econometrics. Week 4. Fall Institute of Economic Studies Faculty of Social Sciences Charles University in Prague

Econometrics. Week 4. Fall Institute of Economic Studies Faculty of Social Sciences Charles University in Prague Econometrics Week 4 Institute of Economic Studies Faculty of Social Sciences Charles University in Prague Fall 2012 1 / 23 Recommended Reading For the today Serial correlation and heteroskedasticity in

More information

STAT 328 (Statistical Packages)

STAT 328 (Statistical Packages) Department of Statistics and Operations Research College of Science King Saud University Exercises STAT 328 (Statistical Packages) nashmiah r.alshammari ^-^ Excel and Minitab - 1 - Write the commands of

More information

Practical Solutions to Behrens-Fisher Problem: Bootstrapping, Permutation, Dudewicz-Ahmed Method

Practical Solutions to Behrens-Fisher Problem: Bootstrapping, Permutation, Dudewicz-Ahmed Method Practical Solutions to Behrens-Fisher Problem: Bootstrapping, Permutation, Dudewicz-Ahmed Method MAT653 Final Project Yanjun Yan Syracuse University Nov. 22, 2005 Outline Outline 1 Introduction 2 Problem

More information

p(z)

p(z) Chapter Statistics. Introduction This lecture is a quick review of basic statistical concepts; probabilities, mean, variance, covariance, correlation, linear regression, probability density functions and

More information

Robust Outcome Analysis for Observational Studies Designed Using Propensity Score Matching

Robust Outcome Analysis for Observational Studies Designed Using Propensity Score Matching The work of Kosten and McKean was partially supported by NIAAA Grant 1R21AA017906-01A1 Robust Outcome Analysis for Observational Studies Designed Using Propensity Score Matching Bradley E. Huitema Western

More information

PRINCIPLES OF STATISTICAL INFERENCE

PRINCIPLES OF STATISTICAL INFERENCE Advanced Series on Statistical Science & Applied Probability PRINCIPLES OF STATISTICAL INFERENCE from a Neo-Fisherian Perspective Luigi Pace Department of Statistics University ofudine, Italy Alessandra

More information

The formal relationship between analytic and bootstrap approaches to parametric inference

The formal relationship between analytic and bootstrap approaches to parametric inference The formal relationship between analytic and bootstrap approaches to parametric inference T.J. DiCiccio Cornell University, Ithaca, NY 14853, U.S.A. T.A. Kuffner Washington University in St. Louis, St.

More information

IV Estimation and its Limitations: Weak Instruments and Weakly Endogeneous Regressors

IV Estimation and its Limitations: Weak Instruments and Weakly Endogeneous Regressors IV Estimation and its Limitations: Weak Instruments and Weakly Endogeneous Regressors Laura Mayoral IAE, Barcelona GSE and University of Gothenburg Gothenburg, May 2015 Roadmap Deviations from the standard

More information

(3) (S) THE BIAS AND STABILITY OF JACK -KNIFE VARIANCE ESTIMATOR IN RATIO ESTIMATION

(3) (S) THE BIAS AND STABILITY OF JACK -KNIFE VARIANCE ESTIMATOR IN RATIO ESTIMATION THE BIAS AND STABILITY OF JACK -KNIFE VARIANCE ESTIMATOR IN RATIO ESTIMATION R.P. Chakrabarty and J.N.K. Rao University of Georgia and Texas A M University Summary The Jack -Knife variance estimator v(r)

More information

Bivariate Paired Numerical Data

Bivariate Paired Numerical Data Bivariate Paired Numerical Data Pearson s correlation, Spearman s ρ and Kendall s τ, tests of independence University of California, San Diego Instructor: Ery Arias-Castro http://math.ucsd.edu/~eariasca/teaching.html

More information

GROUPED SURVIVAL DATA. Florida State University and Medical College of Wisconsin

GROUPED SURVIVAL DATA. Florida State University and Medical College of Wisconsin FITTING COX'S PROPORTIONAL HAZARDS MODEL USING GROUPED SURVIVAL DATA Ian W. McKeague and Mei-Jie Zhang Florida State University and Medical College of Wisconsin Cox's proportional hazard model is often

More information

A consideration of the chi-square test of Hardy-Weinberg equilibrium in a non-multinomial situation

A consideration of the chi-square test of Hardy-Weinberg equilibrium in a non-multinomial situation Ann. Hum. Genet., Lond. (1975), 39, 141 Printed in Great Britain 141 A consideration of the chi-square test of Hardy-Weinberg equilibrium in a non-multinomial situation BY CHARLES F. SING AND EDWARD D.

More information

Bootstrap Tests: How Many Bootstraps?

Bootstrap Tests: How Many Bootstraps? Bootstrap Tests: How Many Bootstraps? Russell Davidson James G. MacKinnon GREQAM Department of Economics Centre de la Vieille Charité Queen s University 2 rue de la Charité Kingston, Ontario, Canada 13002

More information

A3. Statistical Inference

A3. Statistical Inference Appendi / A3. Statistical Inference / Mean, One Sample-1 A3. Statistical Inference Population Mean μ of a Random Variable with known standard deviation σ, and random sample of size n 1 Before selecting

More information

A Test of Homogeneity Against Umbrella Scale Alternative Based on Gini s Mean Difference

A Test of Homogeneity Against Umbrella Scale Alternative Based on Gini s Mean Difference J. Stat. Appl. Pro. 2, No. 2, 145-154 (2013) 145 Journal of Statistics Applications & Probability An International Journal http://dx.doi.org/10.12785/jsap/020207 A Test of Homogeneity Against Umbrella

More information

Solar neutrinos are the only known particles to reach Earth directly from the solar core and thus allow to test directly the theories of stellar evolu

Solar neutrinos are the only known particles to reach Earth directly from the solar core and thus allow to test directly the theories of stellar evolu Absence of Correlation between the Solar Neutrino Flux and the Sunspot Number Guenther Walther Dept. of Statistics, Stanford University, Stanford, CA 94305 Abstract There exists a considerable amount of

More information

Prentice Hall Stats: Modeling the World 2004 (Bock) Correlated to: National Advanced Placement (AP) Statistics Course Outline (Grades 9-12)

Prentice Hall Stats: Modeling the World 2004 (Bock) Correlated to: National Advanced Placement (AP) Statistics Course Outline (Grades 9-12) National Advanced Placement (AP) Statistics Course Outline (Grades 9-12) Following is an outline of the major topics covered by the AP Statistics Examination. The ordering here is intended to define the

More information

Specification Tests for Families of Discrete Distributions with Applications to Insurance Claims Data

Specification Tests for Families of Discrete Distributions with Applications to Insurance Claims Data Journal of Data Science 18(2018), 129-146 Specification Tests for Families of Discrete Distributions with Applications to Insurance Claims Data Yue Fang China Europe International Business School, Shanghai,

More information

BTRY 4090: Spring 2009 Theory of Statistics

BTRY 4090: Spring 2009 Theory of Statistics BTRY 4090: Spring 2009 Theory of Statistics Guozhang Wang September 25, 2010 1 Review of Probability We begin with a real example of using probability to solve computationally intensive (or infeasible)

More information

Diagnostic Test for GARCH Models Based on Absolute Residual Autocorrelations

Diagnostic Test for GARCH Models Based on Absolute Residual Autocorrelations Diagnostic Test for GARCH Models Based on Absolute Residual Autocorrelations Farhat Iqbal Department of Statistics, University of Balochistan Quetta-Pakistan farhatiqb@gmail.com Abstract In this paper

More information

Hypothesis Testing for Var-Cov Components

Hypothesis Testing for Var-Cov Components Hypothesis Testing for Var-Cov Components When the specification of coefficients as fixed, random or non-randomly varying is considered, a null hypothesis of the form is considered, where Additional output

More information

Business Statistics. Lecture 10: Course Review

Business Statistics. Lecture 10: Course Review Business Statistics Lecture 10: Course Review 1 Descriptive Statistics for Continuous Data Numerical Summaries Location: mean, median Spread or variability: variance, standard deviation, range, percentiles,

More information

The Adequate Bootstrap

The Adequate Bootstrap The Adequate Bootstrap arxiv:1608.05913v1 [stat.me] 21 Aug 2016 Toby Kenney Department of Mathematics and Statistics, Dalhousie University and Hong Gu Department of Mathematics and Statistics, Dalhousie

More information

Two Measurement Procedures

Two Measurement Procedures Test of the Hypothesis That the Intraclass Reliability Coefficient is the Same for Two Measurement Procedures Yousef M. Alsawalmeh, Yarmouk University Leonard S. Feldt, University of lowa An approximate

More information

Stat 231 Exam 2 Fall 2013

Stat 231 Exam 2 Fall 2013 Stat 231 Exam 2 Fall 2013 I have neither given nor received unauthorized assistance on this exam. Name Signed Date Name Printed 1 1. Some IE 361 students worked with a manufacturer on quantifying the capability

More information

Understanding Ding s Apparent Paradox

Understanding Ding s Apparent Paradox Submitted to Statistical Science Understanding Ding s Apparent Paradox Peter M. Aronow and Molly R. Offer-Westort Yale University 1. INTRODUCTION We are grateful for the opportunity to comment on A Paradox

More information

Highly Robust Variogram Estimation 1. Marc G. Genton 2

Highly Robust Variogram Estimation 1. Marc G. Genton 2 Mathematical Geology, Vol. 30, No. 2, 1998 Highly Robust Variogram Estimation 1 Marc G. Genton 2 The classical variogram estimator proposed by Matheron is not robust against outliers in the data, nor is

More information

A Bootstrap Test for Causality with Endogenous Lag Length Choice. - theory and application in finance

A Bootstrap Test for Causality with Endogenous Lag Length Choice. - theory and application in finance CESIS Electronic Working Paper Series Paper No. 223 A Bootstrap Test for Causality with Endogenous Lag Length Choice - theory and application in finance R. Scott Hacker and Abdulnasser Hatemi-J April 200

More information

NONPARAMETRICS. Statistical Methods Based on Ranks E. L. LEHMANN HOLDEN-DAY, INC. McGRAW-HILL INTERNATIONAL BOOK COMPANY

NONPARAMETRICS. Statistical Methods Based on Ranks E. L. LEHMANN HOLDEN-DAY, INC. McGRAW-HILL INTERNATIONAL BOOK COMPANY NONPARAMETRICS Statistical Methods Based on Ranks E. L. LEHMANN University of California, Berkeley With the special assistance of H. J. M. D'ABRERA University of California, Berkeley HOLDEN-DAY, INC. San

More information

Introduction to Statistical Data Analysis III

Introduction to Statistical Data Analysis III Introduction to Statistical Data Analysis III JULY 2011 Afsaneh Yazdani Preface Major branches of Statistics: - Descriptive Statistics - Inferential Statistics Preface What is Inferential Statistics? The

More information

Glossary. The ISI glossary of statistical terms provides definitions in a number of different languages:

Glossary. The ISI glossary of statistical terms provides definitions in a number of different languages: Glossary The ISI glossary of statistical terms provides definitions in a number of different languages: http://isi.cbs.nl/glossary/index.htm Adjusted r 2 Adjusted R squared measures the proportion of the

More information

2008 Winton. Statistical Testing of RNGs

2008 Winton. Statistical Testing of RNGs 1 Statistical Testing of RNGs Criteria for Randomness For a sequence of numbers to be considered a sequence of randomly acquired numbers, it must have two basic statistical properties: Uniformly distributed

More information

ADJUSTED POWER ESTIMATES IN. Ji Zhang. Biostatistics and Research Data Systems. Merck Research Laboratories. Rahway, NJ

ADJUSTED POWER ESTIMATES IN. Ji Zhang. Biostatistics and Research Data Systems. Merck Research Laboratories. Rahway, NJ ADJUSTED POWER ESTIMATES IN MONTE CARLO EXPERIMENTS Ji Zhang Biostatistics and Research Data Systems Merck Research Laboratories Rahway, NJ 07065-0914 and Dennis D. Boos Department of Statistics, North

More information

Simulating Properties of the Likelihood Ratio Test for a Unit Root in an Explosive Second Order Autoregression

Simulating Properties of the Likelihood Ratio Test for a Unit Root in an Explosive Second Order Autoregression Simulating Properties of the Likelihood Ratio est for a Unit Root in an Explosive Second Order Autoregression Bent Nielsen Nuffield College, University of Oxford J James Reade St Cross College, University

More information