Constructive Cross-Validation in Linear Prediction

Approaches for regressor construction in the linear prediction problem are investigated in a framework similar to partial least squares and continuum regression, but weighted to allow for custom specification of an evaluative scheme. A cross-validatory continuum regression procedure is proposed, and shown to compare well with ordinary continuum regression in empirical demonstrations.     

Cross validation of ridge regression estimator in autocorrelated linear regression models

ABSTRACT

In this paper, we investigated the cross validation measures, namely OCV, GCV and Cp under the linear regression models when the error structure is autocorrelated and regressor data are correlated. The best performed ridge regression estimator is obtained by getting the optimal ridge parameter so as to minimize these measures. A Monte Carlo simulation study is given to see how the optimal ridge parameter is affected by autocorrelation and the strength of multicollinearity.     

A Note on a Commonly Used Ridge Regression Monte Carlo Design

Ridge estimators are usually examined through Monte Carlo simulations since their properties are difficult to obtain analytically. In this paper we argue that a simulation design commonly used in the literature will give biased results of Monte Carlo simulations in favor of ridge regression over ordinary least square estimators. Specifically, it is argued that the properties of ridge estimators that are functions of p distinct regressor eigenvalues should not be evaluated through Monte Carlo designs using only two distinct eigenvalues.     

Error variance estimation via least squares for small sample nonparametric regression

In this paper we explore statistical properties of some difference-based approaches to estimate an error variance for small sample based on nonparametric regression which satisfies Lipschitz condition. Our study is motivated by Tong and Wang (2005), who estimated error variance using a least squares approach. They considered the error variance as the intercept in a simple linear regression which was obtained from the expectation of their lag-k Rice estimator. Their variance estimators are highly dependent on the setting of a regressor and weight of their simple linear regression. Although this regressor and weight can be varied based on the characteristic of an unknown nonparametric mean function, Tong and Wang (2005) have used a fixed regressor and weight in a large sample and gave no indication of how to determine the regressor and the weight. In this paper, we propose a new approach via local quadratic approximation to determine this regressor and weight. Using our proposed regressor and weight, we estimate the error variance as the intercept of simple linear regression using both ordinary least squares and weighted least squares. Our approach applies to both small and large samples, while most existing difference-based methods are appropriate solely for large samples. We compare the performance of our approach with other existing approaches using extensive simulation study. The advantage of our approach is demonstrated using a real data set.     

Ridge regression estimators in the linear regression models with non-spherical errors

The present paper considers a family of ordinary ridge regression estimators in the linear regression model when the disturbances covariance matrix depends upon a few unknown parameters. An asymptotic expansion for the distribution of the ridge regression estimator is developed and under the quadratic loss function its asymptotic risk is compared with that of the feasible GLS estimator.     

Multivariate Calibration — Direct and Indirect Regression Methodology

This paper tries first to introduce and motivate the methodology of multivariate calibration. Next a review is given, mostly avoiding technicalities, of the somewhat messy theory of the subject. Two approaches are distinguished: the estimation approach (controlled calibration) and the prediction approach (natural calibration). Among problems discussed are the choice of estimator, the choice of confidence region, methodology for handling situations with more variables than observations, near-collinearities (with counter-measures like ridge type regression, principal components regression, partial least squares regression and continuum regression), pretreatment of data, and cross-validation vs true prediction. Examples discussed in detail concern estimation of the age of a rhinoceros from its horn lengths (low-dimensional), and nitrate prediction in waste-water from high-dimensional spectroscopic measurements.     

On the almost unbiased ridge regression estimator

The purpose of this paper is two-fold. One is to compare the almost unbiased generalized ridge regression (AUGRR) estimator proposed by Singh, Chaubey and Dwivedi (1986) with the generalized ridge regression (GRR) estimator and with the ordinary least squares (OLS) estimator in terms of the mean squared error criterion. Second is to examine small sample properties of the operational almost unbiased ordinary ridge regression (AUORR) estimator by Monte Carlo experiments.     

Regressor space outliers in ridge regression

A simple analytical expression is derived for leverage in ridge regression. Leverage is shown to be a monotonically decreasing function of the value of the ridge parameter. This reduction in leverage is greatest for those observations lying substantially in the direction of the minor principal axes. Thus, ridge estimation copes with outliers in regressor space by downweighting their influence. A brief illustration is provided.     

Polynomial regression with errors in the variables

A polynomial functional relationship with errors in both variables can be consistently estimated by constructing an ordinary least squares estimator for the regression coefficients, assuming hypothetically the latent true regressor variable to be known, and then adjusting for the errors. If normality of the error variables can be assumed, the estimator can be simplified considerably. Only the variance of the errors in the regressor variable and its covariance with the errors of the response variable need to be known. If the variance of the errors in the dependent variable is also known, another estimator can be constructed.     

Comment on hoerl and kennard's ridge regression simulation methodology

Ridge regression has received strong support in several Monte carlo studies, leading some authors to advocate its general use. It is argued, however, that these studies were strongly biased in favor of ridge regression by simulating regression coefficient vectors centered at the origin; a condition well suited for a shrinkage technique. Studies which modeled some non-zero regression coefficients and which showed only qualified support for ridge regression are cited in support of this argument. It is concluded that only to the extent that ridge regression type coefficient vectors actually underlie real data sets -a poorly understood phenomenon - will ridge regression be of use.     

A Nonparametric Test of the Predictive Regression Model

This article considers testing the significance of a regressor with a near unit root in a predictive regression model. The procedures discussed in this article are nonparametric, so one can test the significance of a regressor without specifying a functional form. The results are used to test the null hypothesis that the entire function takes the value of zero. We show that the standardized test has a normal distribution regardless of whether there is a near unit root in the regressor. This is in contrast to tests based on linear regression for this model where tests have a nonstandard limiting distribution that depends on nuisance parameters. Our results have practical implications in testing the significance of a regressor since there is no need to conduct pretests for a unit root in the regressor and the same procedure can be used if the regressor has a unit root or not. A Monte Carlo experiment explores the performance of the test for various levels of persistence of the regressors and for various linear and nonlinear alternatives. The test has superior performance against certain nonlinear alternatives. An application of the test applied to stock returns shows how the test can improve inference about predictability.     

Reducing errors-in-variables bias in linear regression using compact genetic algorithms

A new technique is devised to mitigate the errors-in-variables bias in linear regression. The procedure mimics a 2-stage least squares procedure where an auxiliary regression which generates a better behaved predictor variable is derived. The generated variable is then used as a substitute for the error-prone variable in the first-stage model. The performance of the algorithm is tested by simulation and regression analyses. Simulations suggest the algorithm efficiently captures the additive error term used to contaminate the artificial variables. Regressions provide further credit to the simulations as they clearly show that the compact genetic algorithm-based estimate of the true but unobserved regressor yields considerably better results. These conclusions are robust across different sample sizes and different variance structures imposed on both the measurement error and regression disturbances.     

Estimating the structural dimension of regressions via parametric inverse regression

A new estimation method for the dimension of a regression at the outset of an analysis is proposed. A linear subspace spanned by projections of the regressor vector X , which contains part or all of the modelling information for the regression of a vector Y on X , and its dimension are estimated via the means of parametric inverse regression. Smooth parametric curves are fitted to the p inverse regressions via a multivariate linear model. No restrictions are placed on the distribution of the regressors. The estimate of the dimension of the regression is based on optimal estimation procedures. A simulation study shows the method to be more powerful than sliced inverse regression in some situations.     

Fractional principal components regression: a general approach to biased estimators

Several biased estimators have been proposed as alternatives to the least squares estimator when multicollinearity is present in the multiple linear regression model. The ridge estimator and the principal components estimator are two techniques that have been proposed for such problems. In this paper the class of fractional principal component estimators is developed for the multiple linear regression model. This class contains many of the biased estimators commonly used to combat multicollinearity. In the fractional principal components framework, two new estimation techniques are introduced. The theoretical performances of the new estimators are evaluated and their small sample properties are compared via simulation with the ridge, generalized ridge and principal components estimators     

A simulation study of deterministic ridge estimators

Some deterministic ridge rules are proposed and their finite sample properties are studied. Further, a simulation study is also conducted. Based on the simulation results, the proposed ridge estimators can improve the mean squared error over the least squares estimator, provided that the condition number of correlation matrices in the regression model is large, say at least 1,000.     

A comparison of some lack of fit tests based on near replicates

Several tests for regression lack of fit proposed by Christensen (1989), Shillington (1979) and Neill and Johnson (1985) are compared. The tests considered are applicable for the case of nonreplication and reduce to the classical lack of fit test when independent replications are available. A simulation study is used to compare the size and power of the test procedures for small sample sizes and various configurations of nonreplication. In addition, each test is shown to be consistent as well as invariant with respect to location and scale changes made on the regressor variables.     

Effects of correlated disturbances of some regression problems

In a linear regression model the disturbances are assumed to be independently distributed. If the correlation among the disturbances exists, then the usual F statistics have not the F distribution and the distributions depend on the regressor variables. This paper gives bounds on the F statistics, whose distributions do not depends on the regressor variables. The bounds are applied to a test on a general linear hypothesis of the regression coefficients and to evaluate the confidence level of a prediction set.     

Selection of the Ridge Parameter Using Mathematical Programming

Ridge regression solves multicollinearity problems by introducing a biasing parameter that is called ridge parameter; it shrinks the estimates and their standard errors in order to reach acceptable results. Selection of the ridge parameter was done using several subjective and objective techniques that are concerned with certain criteria. In this study, selection of the ridge parameter depends on other important statistical measures to reach a better value of the ridge parameter. The proposed ridge parameter selection technique depends on a mathematical programming model and the results are evaluated using a simulation study. The performance of the proposed method is good when the error variance is greater than or equal to one; the sample consists of 20 observations, the number of explanatory variables in the model is 2, and there is a very strong correlation between the two explanatory variables.     

To be or not to be valid in testing the significance of the slope in simple quantitative linear models with autocorrelated errors

In this article, the validity of procedures for testing the significance of the slope in quantitative linear models with one explanatory variable and first-order autoregressive [AR(1)] errors is analyzed in a Monte Carlo study conducted in the time domain. Two cases are considered for the regressor: fixed and trended versus random and AR(1). In addition to the classical t -test using the Ordinary Least Squares (OLS) estimator of the slope and its standard error, we consider seven t -tests with n-2\,\hbox{df} built on the Generalized Least Squares (GLS) estimator or an estimated GLS estimator, three variants of the classical t -test with different variances of the OLS estimator, two asymptotic tests built on the Maximum Likelihood (ML) estimator, the F -test for fixed effects based on the Restricted Maximum Likelihood (REML) estimator in the mixed-model approach, two t -tests with n - 2 df based on first differences (FD) and first-difference ratios (FDR), and four modified t -tests using various corrections of the number of degrees of freedom. The FDR t -test, the REML F -test and the modified t -test using Dutilleul's effective sample size are the most valid among the testing procedures that do not assume the complete knowledge of the covariance matrix of the errors. However, modified t -tests are not applicable and the FDR t -test suffers from a lack of power when the regressor is fixed and trended ( i.e. , FDR is the same as FD in this case when observations are equally spaced), whereas the REML algorithm fails to converge at small sample sizes. The classical t -test is valid when the regressor is fixed and trended and autocorrelation among errors is predominantly negative, and when the regressor is random and AR(1), like the errors, and autocorrelation is moderately negative or positive. We discuss the results graphically, in terms of the circularity condition defined in repeated measures ANOVA and of the effective sample size used in correlation analysis with autocorrelated sample data. An example with environmental data is presented.     

A new modified Jackknifed estimator for the Poisson regression model

The Poisson regression is very popular in applied researches when analyzing the count data. However, multicollinearity problem arises for the Poisson regression model when the independent variables are highly intercorrelated. Shrinkage estimator is a commonly applied solution to the general problem caused by multicollinearity. Recently, the ridge regression (RR) estimators and some methods for estimating the ridge parameter k in the Poisson regression have been proposed. It has been found that some estimators are better than the commonly used maximum-likelihood (ML) estimator and some other RR estimators. In this study, the modified Jackknifed Poisson ridge regression (MJPR) estimator is proposed to remedy the multicollinearity. A simulation study and a real data example are provided to evaluate the performance of estimators. Both mean-squared error and the percentage relative error are considered as the performance criteria. The simulation study and the real data example results show that the proposed MJPR method outperforms the Poisson ridge regression, Jackknifed Poisson ridge regression and the ML in all of the different situations evaluated in this paper.