Subset selection in two-factor experiments using randomization restricted designs

This paper studies subset selection procedures for screening in two-factor treatment designs that employ either a split-plot or strip-plot randomization restricted experimental design laid out in blocks. The goal is to select a subset of treatment combinations associated with the largest mean. In the split-plot design, it is assumed that the block effects, the confounding effects (whole-plot error) and the measurement errors are normally distributed. None of the selection procedures developed depend on the block variances. Subset selection procedures are given for both the case of additive and non-additive factors and for a variety of circumstances concerning the confounding effect and measurement error variances. In particular, procedures are given for (1) known confounding effect and measurement error variances (2) unknown measurement error variance but known confounding effect (3) unknown confounding effect and measurement error variances. The constants required to implement the procedures are shown to be obtainable from available FORTRAN programs and tables. Generalization to the case of strip-plot randomization restriction is considered.     

A graphical method for testing the equality of several variances

SUMMARY The problem of testing the equality of several variances arises in many areas. For testing the equality of variances, several tests are available in the literature which demonstrate only the statistical significance of the variances. In this paper, a graphical method is presented for testing the equality of variances. This method simultaneously demonstrates the statistical and engineering significance. Two examples are given to illustrate the proposed graphical method, and the conclusions obtained are compared with the existing tests.     

Selecting the normal population with the largest mean when the variances are bounded

A multiple decision approach to the problem of selecting the population with the largest mean was formulated by Bechhofer (1954), where a single-sample solution was presented for the case of normal populations with known variances. In this paper the problem of selecting the normal population with the largest mean is considered when the population variances are unequal and unknown but are constrained only to be less than a specified upper bound. It is demonstrated that a slight modification of Bechhofer' s procedure will suffice to ensure the probability requirements under this simple constraint for cases of practical interest.     

Some approaches to statistical analysis of a weather modification experiment

Data from a weather modification experiment are examined and a number of statistical analyses reported. The validity of earlier inferences is studied as are the utilities of various statistical methods. The experiment is described. The original analysis of North American Weather Consultants, who conducted the experiment, is reviewed. Data summarization is reported. A major approach to analysis is through the use of cloud-physics covari-ates in regression analyses. Finally, a multivariate analysis is discussed. It appears that the covariates may have been affected by treatment (cloud seeding) and that their use is invalid, not only reducing error variances but removing treatment effect. Some recommendations for improved design of similar future experiments are given in a concluding section, including preliminary trial use of blocking by storms.     

Interval Estimation for the Difference of Two Independent Variances

Analytical methods for interval estimation of differences between variances have not been described. A simple analytical method is given for interval estimation of the difference between variances of two independent samples. It is shown, using simulations, that confidence intervals generated with this method have close to nominal coverage even when sample sizes are small and unequal and observations are highly skewed and leptokurtic, provided the difference in variances is not very large. The method is also adapted for testing the hypothesis of no difference between variances. The test is robust but slightly less powerful than Bonett's test with small samples.     

Empirical likelihood for heteroscedastic partially linear errors-in-variables model with α-mixing errors

In this paper, we apply the empirical likelihood method to heteroscedastic partially linear errors-in-variables model. For the cases of known and unknown error variances, the two different empirical log-likelihood ratios for the parameter of interest are constructed. If the error variances are known, the empirical log-likelihood ratio is proved to be asymptotic chi-square distribution under the assumption that the errors are given by a sequence of stationary α-mixing random variables. Furthermore, if the error variances are unknown, we show that the proposed statistic is asymptotically standard chi-square distribution when the errors are independent. Simulations are carried out to assess the performance of the proposed method.     

Posterior likelihood ratios for evaluation of forensic trace evidence given a two-level model on the data

In forensic science, in order to determine whether sets of traces are from the same source or not, it is widely advocated to evaluate evidential value of similarity of the traces by likelihood ratios (LRs). If traces are expressed by measurements following a two-level model with random effects and known variances, closed LR formulas are available given normality, or kernel density distributions, on the effects. For the known variances estimators are used though, which leads to uncertainty on the resulting LRs which is hard to quantify. The above is analyzed in an approach in which both effects and variances are random, following standard prior distributions on univariate data, leading to posterior LRs. For non-informative and conjugate priors, closed LR formulas are obtained that are interesting in structure and generalize a known result given fixed variance. A semi-conjugate prior on the model seems usable in many applications. It is described how to obtain credible intervals using Monte Carlo Markov Chain and regular simulation, and an example is described for comparison of XTC tablets based on MDMA content. In this way, uncertainty on LR estimation is expressed more clearly which makes the evidential value more transparent in a judicial context.     

Estimation of two order restricted normal means with unknown and possibly unequal variances

Estimation of each of and linear functions of two order restricted normal means is considered when variances are unknown and possibly unequal. We replace unknown variances with sample variances and construct isotonic regression estimators, which we call in our paper the plug-in estimators, to estimate ordered normal means. Under squared error loss, a necessary and sufficient condition is given for the plug-in estimators to improve upon the unrestricted maximum likelihood estimators uniformly. As for the estimation of linear functions of ordered normal means, we also show that when variances are known, the restricted maximum likelihood estimator always improves upon the unrestricted maximum likelihood estimator uniformly, but when variances are unknown, the plug-in estimator does not always improve upon the unrestricted maximum likelihood estimator uniformly.     

Some New Estimators for Small-Area Means With Application to the Assessment of Farmland Values

Regression models that account for main state effects and nested county effects are considered for the assessment of farmland values. Empirical predictors obtained by replacing the unknown variances in the formulas of the optimal predictors by maximum likelihood estimates are presented. The computations are carried out by simple iterations between two SAS procedures. Estimators for the prediction variances are derived, and a modification to secure the robustness of the predictors is proposed. The procedure is applied to data on nonirrigated cropland in the Corn Belt states and is shown to yield predictors with considerably lower prediction mean squared errors than the survey estimators and other regression-type estimators.     

A note on the greenwood variance estimator in cohort life tables: Isolating the effect of one source of bias

A modification of the Greenwood variance estimator is defined and shown to be free of bias whenever its constitu­ent interval estimators are conditionally unbiased, given the sample size at the start of the interval. Using the modified estimator as a standard of comparison, the original Greenwood estimator is seen to have an intrinsic positive bias.Under­estimation of variances through the use of Greenwood's formula must be due to bias in the constituent interval estimators and/or, with fixed interval bounds, due to disregarding the random character of the total number of life table intervals to exhaustion of ttje sample. Some easy to prove properties of the modified and the original Greenwood estimators are stated that apply in the absence of censoring. A suggest­ion is made for reducing the bias of the interval variance estimators.     

Exact and Approximate Distributions of the Maximum Likelihood Estimate in a Simple Factor Analysis Model

We study a factor analysis model with two normally distributed observations and one factor. In the case when the errors have equal variance, the maximum likelihood estimate of the factor loading is given in closed form. Exact and approximate distributions of the maximum likelihood estimate are considered. The exact distribution function is given in a complex form that involves the incomplete Beta function. Approximations to the distribution function are given for the cases of large sample sizes and small error variances. The accuracy of the approximations is discussed     

Robust Two-Sample Statistics for Testing Equality of Means: a Simulation Study

When testing the equality of the means from two independent normally distributed populations given that the variances of the two populations are unknown but assumed equal, the classical two-sample t-test is recommended. If the underlying population distributions are normal with unequal and unknown variances, either Welch's t-statistic or Satterthwaite's Approximate F-test is suggested. However, Welch's procedure is non-robust under most non-normal distributions. There is a variable tolerance level around the strict assumptions of data independence, homogeneity of variances and normality of the distributions. Few textbooks offer alternatives when one or more of the underlying assumptions are not defensible.     

Optimal designs for the development of personalized treatment rules

We study the design of multi-armed parallel group clinical trials to estimate personalized treatment rules that identify the best treatment for a given patient with given covariates. Assuming that the outcomes in each treatment arm are given by a homoscedastic linear model, with possibly different variances between treatment arms, and that the trial subjects form a random sample from an unselected overall population, we optimize the (possibly randomized) treatment allocation allowing the allocation rates to depend on the covariates. We find that, for the case of two treatments, the approximately optimal allocation rule does not depend on the value of the covariates but only on the variances of the responses. In contrast, for the case of three treatments or more, the optimal treatment allocation does depend on the values of the covariates as well as the true regression coefficients. The methods are illustrated with a recently published dietary clinical trial.     

Cramér-Rao lower bounds for estimators based on censored data

The Cramér-Rao lower bounds for the variances of unbiased estimators based on censored data are given. Useful techniques of evaluation are then derived for these lower bounds. Examples are given to illustrate these techniques. Small-sample comparisons are made between the resulting lower bounds, the variances of the best linear unbiased estimators, and the variances of unbiased esti-mators which are based on the maximum likelihood estimators.     

A Comparison of Bayesian Models of Heteroscedasticity in Nested Normal Data

We consider the fitting of a Bayesian model to grouped data in which observations are assumed normally distributed around group means that are themselves normally distributed, and consider several alternatives for accommodating the possibility of heteroscedasticity within the data. We consider the case where the underlying distribution of the variances is unknown, and investigate several candidate prior distributions for those variances. In each case, the parameters of the candidate priors (the hyperparameters) are themselves given uninformative priors (hyperpriors). The most mathematically convenient model for the group variances is to assign them inverse gamma distributed priors, the inverse gamma distribution being the conjugate prior distribution for the unknown variance of a normal population. We demonstrate that for a wide class of underlying distributions of the group variances, a model that assigns the variances an inverse gamma-distributed prior displays favorable goodness-of-fit properties relative to other candidate priors, and hence may be used as standard for modeling such data. This allows us to take advantage of the elegant mathematical property of prior conjugacy in a wide variety of contexts without compromising model fitness. We test our findings on nine real world publicly available datasets from different domains, and on a wide range of artificially generated datasets.     

A thurstone-type model for paired comparisons with unequal numbers of repetitions

In paired comparison experiments t objects are ranked for any particular characteristic x by offering the objects in all possible pairs to a judge, each pair being repeated a certain number of times. The judge is to express his preference by giving a score 1 to the preferred object and a score 0 to the non-preferred object. A modification of Thurstone Model for analysis of data from such experiments has been given by Mosteller ‘1951a,b,c’. In this paper angular transformation is used to generalize Mosteller1s model in order to make the preference proportions independent and incidentally ensure homo-scedastlcity of variances and correlations and additivlty of scale in the subjective continuum for the stimuli." The model is extended to unequal numbers of repetitions of the pairs. Using the model two different types of treatment ratings are obtained along with the respective standard errors for moderately large numbers of repetitions, one setting the location parameter S, "0 and the other using the constraint S1 + S2 +…,+ St = 0.     

Optimal bounded influence regression and scale m-estimators in the context of experimental desing

The problem of simultaneous robust estimation of regression and scale parameters in the linear regression model is studied in the context of experimental design. Optimal M-estimates are given for a modified optimization problem of minimizing the asymptotic variances under bounded influence functions. This is done by reducing the multidimensional regression problem to the problem of estimating one-dimensional location and scale. For the location-scale case two subfamilies of optimal score functions are described in detail along with comparisons of the asymptotic variances and gross-error-sensitivities of the corresponding M-estimators. It turns out that, even for small gross-error-sensitivities, one of the subfamilies provides variances which are close to those of the nonrobust maximum likelihood estimators.     

Prediction and forecasting in linear models with measurement error

We consider measurement error models within the time series unobserved component framework. A variable of interest is observed with some measurement error and modelled as an unobserved component. The forecast and the prediction of this variable given the observed values is given by the Kalman filter and smoother along with their conditional variances. By expressing the forecasts and predictions as weighted averages of the observed values, we investigate the effect of estimation error in the measurement and observation noise variances. We also develop corrected standard errors for prediction and forecasting accounting for the fact that the measurement and observation error variances are estimated by the same sample that is used for forecasting and prediction purposes. We apply the theory to the Yellowstone grizzly bears and US index of production datasets.     

On two-stage selection procedures and related probability-inequalities

In this paper we discuss a modification of the Dudewicz-Dalal procedure for the problem of selecting the population with the largest mean from k normal populations with unknown variances. We derive some inequalities and use them to lower-bound the probability of correct selection. These bounds are applied to the determination of the second-stage sample size which is required in order to achieve a prescribed probability of correct selection. We discuss the resulting procedure and compare it to that of Dudewicz and Dalai (1975).     

Missing values in replicated Latin squares

Designs based on any number of replicated Latin squares are examined for their robustness against the loss of up to three observations randomly scattered throughout the design. The information matrix for the treatment effects is used to evaluate the average variances of the treatment differences for each design in terms of the number of missing values and the size of the design. The resulting average variances are used to assess the overall robustness of the designs. In general, there are 16 different situations for the case of three missing values when there are at least three Latin square replicates in the design. Algebraic expressions may be determined for all possible configurations, but here the best and worst cases are given in detail. Numerical illustrations are provided for the average variances, relative efficiencies, minimum and maximum variances and the frequency counts, showing the effects of the missing values for a range of design sizes and levels of replication.