Minimax robust designs for field experiments

Experimental designs for field experiments are useful in planning agricultural experiments, environmental studies, etc. Optimal designs depend on the spatial correlation structures of field plots. Without knowing the correlation structures exactly in practice, we can study robust designs. Various neighborhoods of covariance matrices are introduced and discussed. Minimax robust design criteria are proposed, and useful results are derived. The generalized least squares estimator is often more efficient than the least squares estimator if the spatial correlation structure belongs to a small neighborhood of a covariance matrix. Examples are given to compare robust designs with optimal designs. The work was partially supported by research grants from the Natural Science and Engineering Research Council of Canada.     

Efficiency of the OLSE for regressions on two-dimensional grids with sinusoidal regressors and spatially correlated errors

For spatial regressions with sinusoidal surfaces, the ordinary least squares estimator (OLSE) is shown to be asymptotically as efficient as the generalized least squares estimator (GLSE) in that the covariance matrices of the two estimators have the same nontrivial limit under the same normalization.     

Optimality of least squares in the seemingly unrelated regression equation model

In this article the authors have investigated the situations in which the single-equation least squares estimator is identical with the generalized least squares estimator in the seemingly unrelated regression model. The condition obtained turned out to be advantageous from an empirical point of view as it permits one to decide whether to go for a single-equation least squares method or Zellner's method with estimated disturbance variance covariance matrix for estimating the coefficients in the model.     

The asymptotic behaviour of the estimated generalized least squares method in the linear regression model

Abstract  In the linear regression model the generalized least squares (GLS) method is only applicable if the covariance matrix of the errors is known but for a scalar factor. Otherwise an estimator for this matrix has to be used. Then we speak of the estimated generalized least squares (EGLS) method. In this paper the asymptotic behaviour of both methods is compared. Results are applied to some standard models commonly used in econometrics     

Small sample corrections for LTS and MCD

The least trimmed squares estimator and the minimum covariance determinant estimator [6] are frequently used robust estimators of regression and of location and scatter. Consistency factors can be computed for both methods to make the estimators consistent at the normal model. However, for small data sets these factors do not make the estimator unbiased. Based on simulation studies we therefore construct formulas which allow us to compute small sample correction factors for all sample sizes and dimensions without having to carry out any new simulations. We give some examples to illustrate the effect of the correction factor.     

A heteroscedasticity-consistent covariance matrix estimator for time series regressions

This paper provides a covariance matrix estimator for the ordinary least squares coefficients of a linear time series model which is consistent even when the disturbances are heteroscedastic. This estimator does not require a formal model of the heteroscedasticity. One can also obtain a direct test of heteroscedasticity, although Monte Carlo experiments show that it may have low power.     

Mean estimation bias in least squares estimation of autoregressive processes

Estimation of the parameters of an autoregressive process with a mean that is a function of time is considered. Approximate expressions for the bias of the least squares estimator of the autoregressive parameters that is due to estimating the unknown mean function are derived. For the case of a mean function that is a polynomial in time, a reparameterization that isolates the bias is given. Using the approximate expressions, a method of modifying the least squares estimator is proposed. A Monte Carlo study of the second-order autoregressive process is presented. The Monte Carlo results agree well with the approximate theory and, generally speaking, the modified least squares estimators performed better than the least squares estimator. For the second-order process we also considered the empirical properties of the estimated generalized least squares estimator of the mean function and the error made in predicting the process one, two and three periods in the future.     

An application of approximate finite sample results to parameter estimation in a linear errors-in-variables model

In the simple errors-in-variables model the least squares estimator of the slope coefficient is known to be biased towards zero for finite sample size as well as asymptotically. In this paper we suggest a new corrected least squares estimator, where the bias correction is based on approximating the finite sample bias by a lower bound. This estimator is computationally very simple. It is compared with previously proposed corrected least squares estimators, where the correction aims at removing the asymptotic bias or the exact finite sample bias. For each type of corrected least squares estimators we consider the theoretical form, which depends on an unknown parameter, as well as various feasible forms. An analytical comparison of the theoretical estimators is complemented by a Monte Carlo study evaluating the performance of the feasible estimators. The new estimator proposed in this paper proves to be superior with respect to the mean squared error.     

Testing Convergence in Income Distribution*

The generalized method of moments (GMM) estimator is often used to test for convergence in income distribution in a dynamic panel set‐up. We argue that though consistent, the GMM estimator utilizes the sample observations inefficiently. We propose a simple ordinary least squares (OLS) estimator with more efficient use of sample information. Our Monte Carlo study shows that the GMM estimator can be very imprecise and severely biased in finite samples. In contrast, the OLS estimator overcomes these shortcomings.     

Concentration behavior of the penalized least squares estimator

Consider the standard nonparametric regression model and take as estimator the penalized least squares function. In this article, we study the trade‐off between closeness to the true function and complexity penalization of the estimator, where complexity is described by a seminorm on a class of functions. First, we present an exponential concentration inequality revealing the concentration behavior of the trade‐off of the penalized least squares estimator around a nonrandom quantity, where such quantity depends on the problem under consideration. Then, under some conditions and for the proper choice of the tuning parameter, we obtain bounds for this nonrandom quantity. We illustrate our results with some examples that include the smoothing splines estimator.     

Simultaneous optimal estimation in linear mixed models

Simultaneous optimal estimation in linear mixed models is considered. A necessary and sufficient condition is presented for the least squares estimator of the fixed effects and the analysis of variance estimator of the variance components to be of uniformly minimum variance simultaneously in a general variance components model. That is, the matrix obtained by orthogonally projecting the covariance matrix onto the orthogonal complement space of the column space of the design matrix is symmetric, each eigenvalue of the matrix is a linear combinations of the variance components and the number of all distinct eigenvalues of the matrix is equal to the the number of the variance components. Under this condition, uniformly optimal unbiased tests and uniformly most accurate unbiased confidence intervals are constructed for the parameters of interest. A necessary and sufficient condition is also given for the equivalence of several common estimators of variance components. Two examples of their application are given.     

On the relative efficiency of estimators which include the initial observations in the estimation of seemingly unrelated regressions with first-order autoregressive disturbances

The generalized least squares estimator for a seemingly unrelated regressions model with first-order vector autoregressive disturbances is outlined, and its efficiency is compared with that of an approximate generalized least squares estimator which ignores the first observation. A scalar index for the loss of efficiency is developed and applied to a special case where the matrix of autoregressive parameters is diagonal and the regressors are smooth. Also, for a more general model, a Monte Carlo study is used to investigate the relative efficiencies of various estimators. The results suggest that Maeshiro (1980) has overstated the case for the exact generalized least squares estimator, because, in many circumstances, it is only marginally better than the approximate generalized least squares estimator.     

The effect of a Durbin–Watson pretest on confidence intervals in regression

Consider a linear regression model and suppose that our aim is to find a confidence interval for a specified linear combination of the regression parameters. In practice, it is common to perform a Durbin–Watson pretest of the null hypothesis of zero first‐order autocorrelation of the random errors against the alternative hypothesis of positive first‐order autocorrelation. If this null hypothesis is accepted then the confidence interval centered on the ordinary least squares estimator is used; otherwise the confidence interval centered on the feasible generalized least squares estimator is used. For any given design matrix and parameter of interest, we compare the confidence interval resulting from this two‐stage procedure and the confidence interval that is always centered on the feasible generalized least squares estimator, as follows. First, we compare the coverage probability functions of these confidence intervals. Second, we compute the scaled expected length of the confidence interval resulting from the two‐stage procedure, where the scaling is with respect to the expected length of the confidence interval centered on the feasible generalized least squares estimator, with the same minimum coverage probability. These comparisons are used to choose the better confidence interval, prior to any examination of the observed response vector.     

Projection‐Based Inference in Randomised Clinical Trials

Covariate information is often available in randomised clinical trials for each subject prior to treatment assignment and is commonly utilised to make covariate adjustment for baseline characteristics predictive of the outcome in order to increase precision and improve power in the detection of a treatment effect. Motivated by a nonparametric covariance analysis, we study a projection approach to making objective covariate adjustment in randomised clinical trials on the basis of two unbiased estimating functions that decouple the outcome and covariate data. The proposed projection approach extends a weighted least‐squares procedure by projecting one of the estimating functions onto the linear subspace spanned by the other estimating function that is E‐ancillary for the average treatment effect. Compared with the weighted least‐squares method, the projection method allows for objective inference on the average treatment effect by exploiting the treatment specific covariate–outcome associations. The resulting projection‐based estimator of the average treatment effect is asymptotically efficient when the treatment‐specific working regression models are correctly specified and is asymptotically more efficient than other existing competitors when the treatment‐specific working regression models are misspecified. The proposed projection method is illustrated by an analysis of data from an HIV clinical trial. In a simulation study, we show that the proposed projection method compares favourably with its competitors in finite samples.     

Instrumental variables estimators of nonparametric models with discrete endogenous regressors

This paper discusses estimation of nonparametric models whose regressor vectors consist of a vector of exogenous variables and a univariate discrete endogenous regressor with finite support. Both identification and estimators are derived from a transform of the model that evaluates the nonparametric structural function via indicator functions in the support of the discrete regressor. A two-step estimator is proposed where the first step constitutes nonparametric estimation of the instrument and the second step is a nonparametric version of two-stage least squares. Linear functionals of the model are shown to be asymptotically normal, and a consistent estimator of the asymptotic covariance matrix is described. For the binary endogenous regressor case, it is shown that one functional of the model is a conditional (on covariates) local average treatment effect, that permits both unobservable and observable heterogeneity in treatments. Finite sample properties of the estimators from a Monte Carlo simulation study illustrate the practicability of the proposed estimators.     

BALANCED VARIABLE ADDITION IN LINEAR MODELS

This paper studies what happens when we move from a short regression to a long regression in a setting where both regressions are subject to misspecification. In this setup, the least‐squares estimator in the long regression may have larger inconsistency than the least‐squares estimator in the short regression. We provide a simple interpretation for the comparison of the inconsistencies and study under which conditions the additional regressors in the long regression represent a “balanced addition” to the short regression.     

Pre-averaging estimators of the ex-post covariance matrix in noisy diffusion models with non-synchronous data

We show how pre-averaging can be applied to the problem of measuring the ex-post covariance of financial asset returns under microstructure noise and non-synchronous trading. A pre-averaged realised covariance is proposed, and we present an asymptotic theory for this new estimator, which can be configured to possess an optimal convergence rate or to ensure positive semi-definite covariance matrix estimates. We also derive a noise-robust Hayashi–Yoshida estimator that can be implemented on the original data without prior alignment of prices. We uncover the finite sample properties of our estimators with simulations and illustrate their practical use on high-frequency equity data.     

Variable selection via composite quantile regression with dependent errors

We propose composite quantile regression for dependent data, in which the errors are from short‐range dependent and strictly stationary linear processes. Under some regularity conditions, we show that composite quantile estimator enjoys root‐n consistency and asymptotic normality. We investigate the asymptotic relative efficiency of composite quantile estimator to both single‐level quantile regression and least‐squares regression. When the errors have finite variance, the relative efficiency of composite quantile estimator with respect to the least‐squares estimator has a universal lower bound. Under some regularity conditions, the adaptive least absolute shrinkage and selection operator penalty leads to consistent variable selection, and the asymptotic distribution of the non‐zero coefficient is the same as that of the counterparts obtained when the true model is known. We conduct a simulation study and a real data analysis to evaluate the performance of the proposed approach.     

Estimation of regression coefficients after a preliminary test for homoscedasticity

When heteroscedasticity of the variances of disturbances in a regression model is suspected, we perform a preliminary test for homoscedasticity prior to estimation of regression coefficients. According to the result of the pre-test, we use either the ordinary least squares estimator or the two-stage Aitken estimator (2SAE). In this paper, using orthonormal regressors, we derive the mean square error (MSE) of the pre-test estimator and show that the 2SAE is inadmissible when the MSE is used as a criterion. Further, we seek the optimal critical value of the pre-test in the sense of minimizing the average relative risk which is based on the MSE.     

Estimation of limited dependent variable models by ordinary least squares and the method of moments

Under certain conditions, a broad class of qualitative and limited dependent variable models can be consistently estimated by the method of moments using a non-iterative correction to the ordinary least squares estimator, with only a small loss of efficiency compared to maximum likelihood estimation. The class of models is that obtained from a classical multinormal regression by any type of censoring or truncation and includes the tobit, probit, two-limit probit, truncated regression, and some variants of the sample selection models. The paper derives the estimators and their asymptotic covariance matrices.