A relationship between generalized and integrated mean square errors

Generalised Mean squared error is a flexible measure of the adequancy of ? repression estimator. It allows specific characteristics of the regression model and its intended use to be In-corportated in the measure itself. Similarly, integrated mean squared error enables a researcher to stipulate particular regions of interest and wi ighting functions in the assessment of a prediction equation. The appeal of both measures is their ability to allow design or model characteristics to directly influence the evaluation of fitted regression models. In this note an e-quivalence of the two measures is established for correctly specified models.     

ON THE USE OF THE STEIN VARIANCE ESTIMATOR IN THE DOUBLE k-CLASS ESTIMATOR IN REGRESSION

This paper investigates the predictive mean squared error performance of a modified double k-class estimator by incorporating the Stein variance estimator. Recent studies show that the performance of the Stein rule estimator can be improved by using the Stein variance estimator. However, as we demonstrate below, this conclusion does not hold in general for all members of the double k-class estimators. On the other hand, an estimator is found to have smaller predictive mean squared error than the Stein variance-Stein rule estimator, over quite large parts of the parameter space.     

A note on shrinkage factors in two stage testimation

We obtain a simple and natural testimator which has locally, at the parametric point corresponding to the prior knowledge, a smaller mean squared error than any other two stage testimator of a location or a scale parameter of an arbitrary distribution.     

On huntseerger type shrinkage estimator

This paper is concerned with Hintsberger type weighted shrinkage estimator of a parameter when a target value of the same is available. Expressions for the bias and the mean squared error of the estimator are derived. Some results concerning the bias, existence of uniformly minimum mean squared error estimator etc. are proved. For certain c to ices of the weight function, numerical results are presented for the pretest type weighted shrinkage estimator of the mean of normal as well as exponential distributions.     

One-sided trimming in small samples with asymmetric contamination

An estimator for location, given a sample of only four or five observations, is proposed. The underlying distribution on of the sample may (with probability p) be contaminated by an outlier from a rightly-skewed distribution. The estimator minimizes the maximum mean squared error over all values of p. In fact, there exists an estimator which is unbiased in both the outlier - free and extreme-outlier cases, but its mean square error is substantially higher than the mean squared error for the minimax estimator. Mean squared errors for various underlying distributional situations are calculated and compared with those of other location estimators such as the mean and the median.     

MSE performance of the weighted average estimators consisting of shrinkage estimators when each individual regression coefficient is estimated

In this paper, we analytically derive the exact formula for the mean squared error (MSE) of two weighted average (WA) estimators for each individual regression coefficient. Further, we execute numerical evaluations to investigate small sample properties of the WA estimators, and compare the MSE performance of the WA estimators with the other shrinkage estimators and the usual OLS estimator. Our numerical results show that (1) the WA estimators have smaller MSE than the other shrinkage estimators and the OLS estimator over a wide region of parameter space; (2) the range where the relative MSE of the WA estimator is smaller than that of the OLS estimator gets narrower as the number of explanatory variables k increases.     

Exactly optimal sampling designs for processes with a product covariance structure

The author considers the problem of finding exactly optimal sampling designs for estimating a second‐order, centered random process on the basis of finitely many observations. The value of the process at an unsampled point is estimated by the best linear unbiased estimator. A weighted integrated mean squared error or the maximum mean squared error is used to measure the performance of the estimator. The author presents a set of necessary and sufficient conditions for a design to be exactly optimal for processes with a product covariance structure. Expansions of these conditions lead to conditions for asymptotic optimality.     

On a class of shrinkage estimators of vector of regression coefficients the

This paper studies a class of shrinkage estimators of the vector of regression coefficients. The small disturbance approximations for the bias and the mean squared error matrix of the estimator are derived. In the sense of mean squared error, these estimators dominate the least squares estimator and the generalized Stein estimator developed by Hosmane (1988).     

Kernel Survival Function Estimation Based on Doubly Censored Data

This work concerns the estimation of a smooth survival function based on doubly censored data. We establish strong consistency and asymptotic normality for a kernel estimator. Moreover, we also obtain an asymptotic expression for the mean integrated squared error, which yields an optimum bandwidth in terms of readily estimable quantities.     

Selecting a test level for random effects

A method is presented for selecting an a-level to use when testing for group difference in a one-way classification random effects model. The a-level is chosen to make the power of the test equal to .5 when the parameters are such that between group mean square and total mean square are equally good minimum expected squared error estimators of the variance of y the estimator of the mean     

EXACT FINITE-SAMPLE PROPERTIES OF A PRE-TEST ESTIMATOR IN RIDGE REGRESSION

This paper discusses a pre-test regression estimator which uses the least squares estimate when it is “large” and a ridge regression estimate for “small” regression coefficients, where the preliminary test is applied separately to each regression coefficient in turn to determine whether it is “large” or “small.” For orthogonal regressors, the exact finite-sample bias and mean squared error of the pre-test estimator are derived. The latter is less biased than a ridge estimator, and over much of the parameter space the pre-test estimator has smaller mean squared error than least squares. A ridge estimator is found to be inferior to the pre-test estimator in terms of mean squared error in many situations, and at worst the latter estimator is only slightly less efficient than the former at commonly used significance levels.     

Interchangeability of the ratio and product methods in sample surveys

The paper demonstrates the interchangeability of the ratio and product methods of estimation i n sample surveys through translati n g the unbiased estimator of the population total of the auxiiart variate (or the study varia te). The values of the translation parameters minimizing the mean squared error are obtained. The allowable departures from this optimum, which still ensure a reduction in the mean squared error, as compared to the traditional case, are indicated.     

Estimating Damages in a Class Action Litigation

In a class action litigation, actual damages are not known exactly and must be estimated. Various estimators are proposed and assessed by using a model that identifies possible sources of error. Estimators that have been used in practice are shown to be seriously biased. An empirical Bayes estimator and an empirical minimal mean squared error estimator are found to be more satisfactory methods for estimating damages.     

Conditional variance estimation via nonparametric generalized additive models

Generalized additive models provide a way of circumventing curse of dimension in a wide range of nonparametric regression problem. In this paper, we present a multiplicative model for conditional variance functions where one can apply a generalized additive regression method. This approach extends Fan and Yao (1998) to multivariate cases with a multiplicative structure. In this approach, we use squared residuals instead of using log-transformed squared residuals. This idea gives a smaller variance than Yu (2017) when the variance of squared error is smaller than the variance of log-transformed squared error. We provide estimators based on quasi-likelihood and an iterative algorithm based on smooth backfitting for generalized additive models. We also provide some asymptotic properties of estimators and the convergence of proposed algorithm. A numerical study shows the empirical evidence of the theory.     

Alternative moment approximations for berkson's minimum logit chi-squared estimator

Expressions are derived for the bias to order J^-1 , the variance to order J^-2 and the mean squared error to order J^-2 of Berkson's minimum logit chi-squared estimator where J is the number of distinct design points. These moment approximations are numerically compared to Monte Carlo estimates of the true moments and the moment approximations of Amemiya (1980) which are appropriate when the “average” number of observations per design point is large. They are used to compare the mean squared error of the minimum logit chi-squared estimator to that of the maximum likelihood estimator and to investigate the effect of bias on confidence intenrals constructed using the minimum logit chi-squared estimator.     

Generalized mean squared error properties of regression estimators

Theobald (1974) compares Ordinary Least Squares and Ridge Regression estimators of regression parameters using a generalized mean squared error criterion. This paper presents the generalized mean squared error of a Principal Components Regression estimator and comparisons are made with each of the above estimators. In general the choice of which estimator to use depends on the magnitude and the orientation of the unknown parameter vector.     

Probability density estimation with data missing at random when covariables are present

This paper addresses the problem of the probability density estimation in the presence of covariates when data are missing at random (MAR). The inverse probability weighted method is used to define a nonparametric and a semiparametric weighted probability density estimators. A regression calibration technique is also used to define an imputed estimator. It is shown that all the estimators are asymptotically normal with the same asymptotic variance as that of the inverse probability weighted estimator with known selection probability function and weights. Also, we establish the mean squared error (MSE) bounds and obtain the MSE convergence rates. A simulation is carried out to assess the proposed estimators in terms of the bias and standard error.     

Mean squared error estimators of small area means using survey weights

Using survey weights, You & Rao [You and Rao, The Canadian Journal of Statistics 2002; 30, 431–439] proposed a pseudo‐empirical best linear unbiased prediction (pseudo‐EBLUP) estimator of a small area mean under a nested error linear regression model. This estimator borrows strength across areas through a linking model, and makes use of survey weights to ensure design consistency and preserve benchmarking property in the sense that the estimators add up to a reliable direct estimator of the mean of a large area covering the small areas. In this article, a second‐order approximation to the mean squared error (MSE) of the pseudo‐EBLUP estimator of a small area mean is derived. Using this approximation, an estimator of MSE that is nearly unbiased is derived; the MSE estimator of You & Rao [You and Rao, The Canadian Journal of Statistics 2002; 30, 431–439] ignored cross‐product terms in the MSE and hence it is biased. Empirical results on the performance of the proposed MSE estimator are also presented. The Canadian Journal of Statistics 38: 598–608; 2010 © 2010 Statistical Society of Canada     

Adaptive wavelet estimation of a density from mixtures under multiplicative censoring

In this paper, a mixture model under multiplicative censoring is considered. We investigate the estimation of a component of the mixture (a density) from the observations. A new adaptive estimator based on wavelets and a hard thresholding rule is constructed for this problem. Under mild assumptions on the model, we study its asymptotic properties by determining an upper bound of the mean integrated squared error over a wide range of Besov balls. We prove that the obtained upper bound is sharp.     

A comparison of estimators of a common correlation coefficient

It is often of interest to test the hypothesis that all off-diagonal elements of the correlation matrix of a multivariate normal distribution are equal. If the hypothesis of equal correlation can be accepted, it then may be of interest to estimate the common correlation coefficient. In this paper, four estimators of the common correlation are compared in terms of bias, variance, mean squared error, adequacy of the normal approximation, and ease of calculation. The average sample correlation is seen to be comparable to the other estimators and is recommended here since it is the easiest to calculate. The estimators are compared using simulation.