Small-Sample Comparisons for Goodness-of-Fit Statistics in One-Way Multinomials with Composite Hypotheses

The small-sample behaviour of power-divergence goodness-of-fit statistics with composite hypotheses was evaluated with multinomial models of up to five cells and up to three parameters. Their performance was assessed by comparing asymptotic test sizes with exact test sizes obtained by enumeration in the near right tail, where 1-?∈?(0.90,?0.95], and in the far right tail, where 1-?∈?(0.95,?0.99]. The study addressed all combinations of power-diparse JAS312HH01.sgmvergence estimates of indices ν?∈?{-1/2,?0,?1/3,?1/2,?2/3,?1,?3/ 2} and power-divergence statistics of indices λ?∈?{-1/2,?0,?1/3,?1/2,?2/3,?1,?3/2}. The results indicate that the asymptotic approximation is sufficiently accurate (by the criterion that the average exact size is no larger than ±10% of the nominal asymptotic test size) in the near right tail when ν=0 and λ=1/2, and in the far right tail when ν=0 and λ=1/3, in both cases providing that the smallest expectation in the composite hypothesis exceeds 5. The only exception to this rule is the case of models that render a near-equiprobable composite hypothesis on the boundaries of the parameter space, where average exact sizes are usually quite different from nominal sizes despite the fact that the smallest expectation in these conditions is usually well above 5.     

Tests of symmetry based on transformed empirical processes

A probability distribution function F is said to be symmetric when 1 ‐ F(x) ‐ F(‐x) = 0 for all x∈ R. Given a sequence of alternatives contiguous to a certain symmetric F₀, the authors are concerned with testing for the null hypothesis of symmetry. The proposed tests are consistent against any nonsymmetric alternative, and their power with respect to the given sequence can easily be optimized. The tests are constructed by means of transformed empirical processes with an adequate selection of the underlying isometry, and the optimum power is obtained by suitably choosing the score functions. The test statistics are very easy to compute and their asymptotic distributions are simple.     

On Statistics Independent of a Sufficient Statistic: Basu's Lemma

The determination of optimal sample sizes for estimating the difference between population means to a desired degree of confidence and precision is a question of economic significance. This question, however, is generally not discussed in statistics texts. Sample sizes to minimize linear sampling costs are proportional to the population standard deviations and inversely proportional to the square roots of the unit sampling costs. Sensitivity analysis shows that the impact of the use of equal rather than optimal sample sizes on the amount of sampling and its cost is not great as long as the unit costs and population variances are comparable.     

Semiparametric Likelihood Based Method for Goodness of Fit Tests and Estimation in Upgraded Mixture Models

We use Owen's (1988, 1990) empirical likelihood method in upgraded mixture models. Two groups of independent observations are available. One is z ₁, ..., z _n which is observed directly from a distribution F ( z ). The other one is x ₁, ..., x _m which is observed indirectly from F ( z ), where the x _i s have density ∫ p ( x | z ) dF ( z ) and p ( x | z ) is a conditional density function. We are interested in testing H ₀: p ( x | z ) = p ( x | z ; θ ), for some specified smooth density function. A semiparametric likelihood ratio based statistic is proposed and it is shown that it converges to a chi-squared distribution. This is a simple method for doing goodness of fit tests, especially when x is a discrete variable with finitely many values. In addition, we discuss estimation of θ and F ( z ) when H ₀ is true. The connection between upgraded mixture models and general estimating equations is pointed out.     

Tests of goodness of fit based on Phi-divergence

In this paper, we introduce a general goodness of fit test based on Phi-divergence. Consistency of the proposed test is established. We then study some special cases of tests for normal, exponential, uniform and Laplace distributions. Through Monte Carlo simulations, the power values of the proposed tests are compared with some known competing tests under various alternatives. Finally, some numerical examples are presented to illustrate the proposed procedure.     

Goodness-of-Fit Tests for Multivariate Distributions

We propose three new statistics, Z _p , C _p , and R _p for testing a p-variate (p ≥ 2) normal distribution and compare them with the prominent test statistics. We show that C _p is overall most powerful and is effective against skew, long-tailed as well as short-tailed symmetric alternatives. We show that Z _p and R _p are most powerful against skew and long-tailed alternatives, respectively. The Z _p and R _p statistics can also be used for testing an assumed p-variate nonnormal distribution.     

A correlation type procedure for assessing multivariate normality

A correlation-type statistic for assessing multivariate normality is described. Its estimated finite sample distribution is tabulated, and its performance against certain alternatives is compared with that of a competing Cramer-von Mises type statistic in a Monte Carlo power study. A set of quadrivariate data is examined as illustration of the procedure.     

Goodness‐of‐Fit Tests for Bivariate and Multivariate Skew‐Normal Distributions

Abstract. Goodness‐of‐fit tests are proposed for the skew‐normal law in arbitrary dimension. In the bivariate case the proposed tests utilize the fact that the moment‐generating function of the skew‐normal variable is quite simple and satisfies a partial differential equation of the first order. This differential equation is estimated from the sample and the test statistic is constructed as an L ₂ ‐type distance measure incorporating this estimate. Extension of the procedure to dimension greater than two is suggested whereas an effective bootstrap procedure is used to study the behaviour of the new method with real and simulated data.     

Chi-Squared Goodness-of-Fit Tests: Cell Selection and Power

To use the Pearson chi-squared statistic to test the fit of a continuous distribution, it is necessary to partition the support of the distribution into k cells. A common practice is to partition the support into cells with equal probabilities. In that case, the power of the chi-squared test may vary substantially with the value of k. The effects of different values of k are investigated with a Monte Carlo power study of goodness-of-fit tests for distributions where location and scale parameters are estimated from the observed data. Allowing for the best choices of k, the Pearson and log-likelihood ratio chi-squared tests are shown to have similar maximum power for wide ranges of alternatives, but this can be substantially less than the power of other well-known goodness-of-fit tests.     

Two-phase regression and goodness of fit

Analysis of two-phase regression has traditionally been carried out using a variety of likelihood approaches. In this paper we present an alternative procedure based on a goodness of fit criterion.

Exact hypothesis tests for a known switch point are developed. Approximate (conservative) tests for an unknown switch point are also obtained     

New Goodness-of-Fit Tests Based on Sample Quantiles

In this article power divergences statistics based on sample quantiles are transformed in order to introduce new goodness-of-fit tests. Quantiles of the distribution of proposed statistics are calculated under uniformity, normality, and exponentiality. Several power comparisons are performed to show that the new tests are generally more powerful than the original ones.     

Bayesian Goodness of Fit Testing with Mixtures of Triangular Distributions

Abstract.  We consider the consistency of the Bayes factor in goodness of fit testing for a parametric family of densities against a non-parametric alternative. Sufficient conditions for consistency of the Bayes factor are determined and demonstrated with priors using certain mixtures of triangular densities.     

Cramér-von Mises statistics for discrete distributions

Cramér-von Mises statistics are developed for use in testing for discrete distributions, and tables are given for tests for the discrete uniform distribution.     

Behaviour of the length test for medium sample sizes

In this note it is shown that even for relatively large sample sites the asymptotic distribution of the smoothed length as derived in Reschenhofer and Bomse (1991) should not be used for the determination of critical values. Therefore extended tables of critical values for both the 1% and 5% levels of significance generated by simulation are presented.     

How Important Is the Effect of Rounding in Goodness-of-Fit

Abstract

In the area of goodness-of-fit there is a clear distinction between the problem of testing the fit of a continuous distribution and that of testing a discrete distribution. In all continuous problems the data is recorded with a limited number of decimals, so in theory one could say that the problem is always of a discrete nature, but it is a common practice to ignore discretization and proceed as if the data is continuous. It is therefore an interesting question whether in a given problem of test of fit, the “limited resolution” in the observed recorded values may be or may be not of concern, if the analysis done ignores this implied discretization. In this article, we address the problem of testing the fit of a continuous distribution with data recorded with a limited resolution. A measure for the degree of discretization is proposed which involves the size of the rounding interval, the dispersion in the underlying distribution and the sample size. This measure is shown to be a key characteristic which allows comparison, in different problems, of the amount of discretization involved. Some asymptotic results are given for the distribution of the EDF (empirical distribution function) statistics that explicitly depend on the above mentioned measure of degree of discretization. The results obtained are illustrated with some simulations for testing normality when the parameters are known and also when they are unknown. The asymptotic distributions are shown to be an accurate approximation for the true finite n distribution obtained by Monte Carlo. A real example from image analysis is also discussed. The conclusion drawn is that in the cases where the value of the measure for the degree of discretization is not “large”, the practice of ignoring discreteness is of no concern. However, when this value is “large”, the effect of ignoring discreteness leads to an exceded number of rejections of the distribution tested, as compared to what would be the number of rejections if no rounding is taking into account. The error made in the number of rejections might be huge.     

A comparison of tests of homogeneity for sparse contingency tables

Five tests of homogeneity for a 2x(k+l) contingency table are compared using Monte Carlo techniques. For these studiesit is assumed that k becomes large in such a way that thecontingency table is sparse for 2xk of the cells, but the sample size in two of the cells remains large. The test statistics studied are: the chi-square approximation to the Pearson test statistic, the chi-square approximation to the likelihood ratio statistic, the normal approximation to Zelterman's (1984)the normal approximation to Pearson's chi-square, and the normal approximation to the likelihood ratio statistic. For the range of parameters studied the chi-square approximation to Pearson's statistic performs consistently well with regard to its size and power.     

Goodness of fit for the inverse Gaussian distribution

For testing the fit of the inverse Gaussian distribution with unknown parameters, the empirical distribution-function statistic A² is studied. Two procedures are followed in constructing the test statistic; they yield the same asymptotic distribution. In the first procedure the parameters in the distribution function are directly estimated, and in the second the distribution function is estimated by its Rao-Blackwell distribution estimator. A table is given for the asymptotic critical points of A². These are shown to depend only on the ratio of the unknown parameters. An analysis is provided of the effect of estimating the ratio to enter the table for A². This analysis enables the proposal of the complete operating procedure, which is sustained by a Monte Carlo study.     

On Goodness-of-Fit Tests for Aalen's Additive Risk Model

Abstract.  In this paper we propose goodness-of-fit tests for Aalen's additive risk model. They are based on test statistics the asymptotic distributions of which are determined under both the null and alternative hypotheses. The results are derived using martingale techniques for counting processes. An important feature of these tests is that they can be adjusted to particular alternatives. One of the alternatives we consider is Cox's multiplicative risk model. It is perhaps remarkable that such a test needs no estimate of the baseline hazard in the Cox model. We present simulation studies which give an impression of the performance of the proposed tests. In addition, the tests are applied to real data sets.