Heteroscedasticity Of Unknown Form Research Articles

Statistical inference for autoregressive models under heteroscedasticity of unknown form

This paper provides an entire inference procedure for the autoregressive model under (conditional) heteroscedasticity of unknown form with a finite variance. We first establish the asymptotic normality of the weighted least absolute deviations estimator (LADE) for the model. Second, we develop the random weighting (RW) method to estimate its asymptotic covariance matrix, leading to the implementation of the Wald test. Third, we construct a portmanteau test for model checking, and use the RW method to obtain its critical values. As a special weighted LADE, the feasible adaptive LADE (ALADE) is proposed and proved to have the same efficiency as its infeasible counterpart. The importance of our entire methodology based on the feasible ALADE is illustrated by simulation results and the real data analysis on three U.S. economic data sets.

Inference in Linear Regression Models with Many Covariates and Heteroscedasticity

ABSTRACTThe linear regression model is widely used in empirical work in economics, statistics, and many other disciplines. Researchers often include many covariates in their linear model specification in an attempt to control for confounders. We give inference methods that allow for many covariates and heteroscedasticity. Our results are obtained using high-dimensional approximations, where the number of included covariates is allowed to grow as fast as the sample size. We find that all of the usual versions of Eicker–White heteroscedasticity consistent standard error estimators for linear models are inconsistent under this asymptotics. We then propose a new heteroscedasticity consistent standard error formula that is fully automatic and robust to both (conditional) heteroscedasticity of unknown form and the inclusion of possibly many covariates. We apply our findings to three settings: parametric linear models with many covariates, linear panel models with many fixed effects, and semiparametric semi-linear models with many technical regressors. Simulation evidence consistent with our theoretical results is provided, and the proposed methods are also illustrated with an empirical application. Supplementary materials for this article are available online.

Automatic Specification Testing for Vector Autoregressions and Multivariate Nonlinear Time Series Models

This article introduces an automatic test for the correct specification of a vector autoregression (VAR) model. The proposed test statistic is a Portmanteau statistic with an automatic selection of the order of the residual serial correlation tested. The test presents several attractive characteristics: simplicity, robustness, and high power in finite samples. The test is simple to implement since the researcher does not need to specify the order of the autocorrelation tested and the proposed critical values are simple to approximate, without resorting to bootstrap procedures. In addition, the test is robust to the presence of conditional heteroscedasticity of unknown form and accounts for estimation uncertainty without requiring the computation of large-dimensional inverses of near-to-singularity covariance matrices. The basic methodology is extended to general nonlinear multivariate time series models. Simulations show that the proposed test presents higher power than the existing ones for models commonly employed in empirical macroeconomics and empirical finance. Finally, the test is applied to the classical bivariate VAR model for GNP (gross national product) and unemployment of Blanchard and Quah (1989) and Evans (1989). Online supplementary material includes proofs and additional details.

Performance of Kibria's Method for the Heteroscedastic Ridge Regression Model: Some Monte Carlo Evidence

It is common for a linear regression model that the error terms display some form of heteroscedasticity and at the same time, the regressors are also linearly correlated. Both of these problems have serious impact on the ordinary least squares (OLS) estimates. In the presence of heteroscedasticity, the OLS estimator becomes inefficient and the similar adverse impact can also be found on the ridge regression estimator that is alternatively used to cope with the problem of multicollinearity. In the available literature, the adaptive estimator has been established to be more efficient than the OLS estimator when there is heteroscedasticity of unknown form. The present article proposes the similar adaptation for the ridge regression setting with an attempt to have more efficient estimator. Our numerical results, based on the Monte Carlo simulations, provide very attractive performance of the proposed estimator in terms of efficiency. Three different existing methods have been used for the selection of biasing parameter. Moreover, three different distributions of the error term have been studied to evaluate the proposed estimator and these are normal, Student's t and F distribution.

Wild Bootstrap Tests for IV Regression

We propose a wild bootstrap procedure for linear regression models estimated by instrumental variables. Like other bootstrap procedures that we proposed elsewhere, it uses efficient estimates of the reduced-form equation(s). Unlike the earlier procedures, it takes account of possible heteroscedasticity of unknown form. We apply this procedure to t tests, including heteroscedasticity-robust t tests, and to the Anderson–Rubin test. We provide simulation evidence that it works far better than older methods, such as the pairs bootstrap. We also show how to obtain reliable confidence intervals by inverting bootstrap tests. An empirical example illustrates the utility of these procedures.

A New Test of Linear Hypotheses in OLS Regression Under Heteroscedasticity of Unknown Form

When the errors in an ordinary least squares (OLS) regression model are heteroscedastic, hypothesis tests involving the regression coefficients can have Type I error rates that are far from the nominal significance level. Asymptotically, this problem can be rectified with the use of a heteroscedasticity-consistent covariance matrix (HCCM) estimator. However, many HCCM estimators do not perform well when the sample size is small or when there exist points of high leverage in the design matrix. Prompted by a connection between MacKinnon and White’s HC2 HCCM estimator and the heterogeneous-variance two-sample t statistic, the authors provide a new statistic for testing linear hypotheses in an OLS regression model that does not assume homoscedasticity. The authors report simulation results showing that their new test maintains better Type I error rate control than existing methods in both the presence and absence of heteroscedasticity.

Simple Robust Testing of Hypotheses in Nonlinear Models

We develop test statistics to test hypotheses in nonlinear weighted regression models with serial correlation or conditional heteroscedasticity of unknown form. The novel aspect is that these tests are simple and do not require the use of heteroscedasticity autocorrelationconsistent (HAC) covariance matrix estimators. This new class of tests uses stochastic transformations to eliminate nuisance parameters as a substitute for consistently estimating the nuisance parameters. We derive the limiting null distributions of these new tests in a general nonlinear setting, and show that although the tests have nonstandard distributions, the distributions depend only on the number of restrictions being tested. We perform some simulations on a simple model and apply the new method of testing to an empirical example and illustrate that the size of the new test is less distorted than tests using HAC covariance matrix estimators.

Two-stage generalized moment method with applications to regressions with heteroscedasticity of unknown form

We describe in a general framework a two-stage generalized moment method. More precisely we explain how to partition the set of estimating constraints and to unfold the set of parameters in order to derive a two-stage approach which is asymptotically equivalent to GMM. Then we consider a linear model Y t = X t b + u t , and the GMM estimator based on the constraints: EX′ t = ,…, EX′ tu t 2p+1 = 0. We explain how to find an asymptotically equivalent estimator by replacing some of the errors u t by their corresponding OLS residuals. Finally, we focus on constraints based on first and third conditional moments. The estimator thus obtained is asymptotically more efficient than the OLS estimator. The efficiency gain is linked with a marginal kurtosis of the u t 's distribution or with the conditional heteroscedasticity.

Estimation in the simple linear regression model when there is heteroscedasticity of unknown form

A well-known problem is that ordinary least squares estimation of the parameters in the usual linear model can be highly ineficient when the error term has a heavy-tailed distribution. Inefficiency is also associated with situations where the error term is heteroscedastic, and standard confidence intervals can have probability coverage substantially different from the nominal level. This paper compares the small-sample efficiency of six methods that address this problem, three of which model the variance heterogeneity nonparametrically. Three methods were found to be relatively ineffective, but the other three perform relatively well. One of the six (M-regression with a Huber φ function and Schweppe weights) was found to have the highest efficiency for most of the situations considered in the simulations, but there might be situations where one of two other methods gives better results. One of these is a new method that uses a running interval smoother to estimate the optimal weights in weighted least squares, and the other is a method recently proposed by Cohen, Dalal, and Tukey. Computing a confidence interval for the slope using a bootstrap technique is also considered.

Patterns in Residential Gas and Electricity Consumption: An Econometric Analysis

In this article, we investigate individual consumption data of natural gas and electricity using modified multiple regression techniques. Because a subset of households have zero gas consumption, the purpose of the study is to ask two empirical questions: 1. How can we quantify potential energy conservation from increasing gas availability? 2. Is it appropriate to estimate and compare electricity demands of gas users and nonusers by the same regression function? We began our investigation based on the methodology of generalized tobit models. Our empirical results indicate that the probit equation for gas connection is exogenous to the amount of potential energy consumption. To answer question 2, we obtain heteroscedasticity-efficient estimates of the electricity-demand equations by a nearest-neighbor estimator to deal with heteroscedasticity of unknown form. The prediction of energy conservation is that electricity demand could be reduced by 2,473.57 kilowatt-hours a year per household if a gas connection is available to nonusers. To reduce electricity demand by this amount, the utility company would need to supply 611.09 Therms of natural gas.