Testing autocorrelation in a system perspective testing autocorrelation

The Breusch-Godfrey test for autocorrelated errors is generalised to cover systems of equations, and the properties of 18 versions of the test are studied using Monte Carlo methods. We show that only one group of tests regularly has actual size close to the nominal size; namely the likelihood ratio tests of the auxiliary regression system that are corrected in some manner for degrees-of-freedom. The Rao Ftest exhibits the best performance, whilst the commonly used TR2 test behaves badly even in single equations. However, the size and power properties of all tests deteriorate sharply as the number of equations increases, the system becomes more dynamic, the exogenous variables become more autocorrelated and the sample size decreases. This performance has, in general, an unknown degree since the interaction amongst these factors does not permit a predictive summary, as might be hoped for by response surface-type approaches.     

A Simulation Study of Some Ridge Regression Estimators under Different Distributional Assumptions

Based on the work of Khalaf and Shukur (2005 Khalaf , G. , Shukur , G. ( 2005 ). Choosing ridge parameters for regression problems . Communications in Statistics – Theory and Methods 34 : 1177 – 1182 .[Taylor & Francis Online], [Web of Science ®] , [Google Scholar]), Alkhamisi et al. (2006 Alkhamisi , M. , Khalaf , G. , Shukur , G. ( 2006 ). Some modifications for choosing ridge parameters . Communications in Statistics – Theory and Methods 35 : 2005 – 2020 .[Taylor & Francis Online], [Web of Science ®] , [Google Scholar]), and Muniz et al. (2010 Muniz , G. , Kibria , B. M. G. , Shukur , G. ( 2010 ). On developing ridge regression parameters: a graphical Investigation. Submitted for Publication . [Google Scholar]), this article considers several estimators for estimating the ridge parameter k. This article differs from aforementioned articles in three ways: (1) Data are generated from Normal, Student's t, and F distributions with appropriate degrees of freedom; (2) The number of regressors considered are from 4–12 instead of 2–4, which are the usual practice; (3) Both mean square error (MSE) and prediction sum of square (PRESS) are considered as the performance criterion. A simulation study has been conducted to compare the performance of the estimators. Based on the simulation study we found that, increasing the correlation between the independent variables has negative effect on the MSE and PRESS. However, increasing the number of regressors has positive effect on MSE and PRESS. When the sample size increases the MSE decreases even when the correlation between the independent variables is large. It is interesting to note that the dominance pictures of the estimators are remained the same under both the MSE and PRESS criterion. However, the performance of the estimators depends on the choice of the assumption of the error distribution of the regression model.     

Performance of Some Ridge Parameters for Probit Regression: With Application to Swedish Job Search Data

In ridge regression, the estimation of the ridge parameter is an important issue. This article generalizes some methods for estimating the ridge parameter for probit ridge regression (PRR) model based on the work of Kibria et al. (2011 Kibria, B. M. G., Månsson, K. and Shukur, G. 2011. Performance of some logistic ridge regression parameters. Computational Economics, DOI: 10.1007/s10614-011-9275-x [Google Scholar]). The performance of these new estimators is judged by calculating the mean squared error (MSE) using Monte Carlo simulations. In the design of the experiment, we chose to vary the sample size and the number of regressors. Furthermore, we generate explanatory variables that are linear combinations of other regressors, which is a common situation in economics. In an empirical application regarding Swedish job search data, we also illustrate the benefits of the new method.     

Seemingly unrelated regressions with covariance matrix of cross-equation ridge regression residuals

Generalized least squares estimation of a system of seemingly unrelated regressions is usually a two-stage method: (1) estimation of cross-equation covariance matrix from ordinary least squares residuals for transforming data, and (2) application of least squares on transformed data. In presence of multicollinearity problem, conventionally ridge regression is applied at stage 2. We investigate the usage of ridge residuals at stage 1, and show analytically that the covariance matrix based on the least squares residuals does not always result in more efficient estimator. A simulation study and an application to a system of firms' gross investment support our finding.     

An illustration of the causality relation between government spending and revenue using wavelet analysis on Finnish data

Quarterly data for the period 1960:1 to 1997:2, conventional tests, a bootstrap simulation approach and a multivariate Rao's F-test have been used to investigate if the causality between government spending and revenue in Finland was changed at the beginning of 1990 due to future plans to create the European Monetary Union (EMU). The results indicate that during the period before 1990, the government revenue Granger-caused spending, while the opposite happened after 1990, which agrees better with Barro's tax smoothing hypothesis. However, when using monthly data instead of quarterly data for almost the same sample period, totally different results have been noted. The general conclusion is that the relationship between spending and revenue in Finland is still not completely understood. The ambiguity of these results may well be due to the fact that there are several time scales involved in the relationship, and that the conventional analyses may be inadequate to separate out the time scale structured relationships between these variables. Therefore, to investigate empirically the relation between these variables we attempt to use the wavelets analysis that enables us to separate out different time scales of variation in the data. We find that time scale decomposition is important for analysing these economic variables.     

Bayesian analysis of a linear mixed model with AR(p) errors via MCMC

We develop Bayesian procedures to make inference about parameters of a statistical design with autocorrelated error terms. Modelling treatment effects can be complex in the presence of other factors such as time; for example in longitudinal data. In this paper, Markov chain Monte Carlo methods (MCMC), the Metropolis-Hastings algorithm and Gibbs sampler are used to facilitate the Bayesian analysis of real life data when the error structure can be expressed as an autoregressive model of order p. We illustrate our analysis with real data.     

The effect of fat-tailed error terms on the properties of systemwise RESET test

The small sample properties of the systemwise RESET (Regression Specification Error Test) test for functional misspecification are investigated using normal and non-normal error terms. When using normally distributed or less heavy tailed error terms, we find the Rao's multivariate F-test to be best among all other alternative test methods (i.e. Wald, Lagrange Multiplier and Likelihood Ratio). Using the bootstrap critical values, however, all test methods perform satisfactorily in almost all situations. However, the test methods perform extremely badly (even the RAO test) when the error terms are very heavy tailed.     

Performance Evaluation Based on the Robust Mahalanobis Distance and Multilevel Modeling Using Two New Strategies

In this article, we propose a general framework for performance evaluation of organizations and individuals over time using routinely collected performance variables or indicators. Such variables or indicators are often correlated over time, with missing observations, and often come from heavy-tailed distributions shaped by outliers. Two new double robust and model-free strategies are used for evaluation (ranking) of sampling units. Strategy 1 can handle missing data using residual maximum likelihood (RML) at stage two, while strategy two handles missing data at stage one. Strategy 2 has the advantage that overcomes the problem of multicollinearity. Strategy one requires independent indicators for the construction of the distances, where strategy two does not. Two different domain examples are used to illustrate the application of the two strategies. Example one considers performance monitoring of gynecologists and example two considers the performance of industrial firms.     

Testing for panel cointegration in an error-correction framework with an application to the Fisher hypothesis

In this article, three innovative panel error-correction model (PECM) tests are proposed. These tests are based on the multivariate versions of the Wald (W), likelihood ratio (LR), and Lagrange multiplier (LM) tests. Using Monte Carlo simulations, the size and power of the tests are investigated when the error terms exhibit both cross-sectional dependence and independence. We find that the LM test is the best option when the error terms follow independent white-noise processes. However, in the more empirically relevant case of cross-sectional dependence, we conclude that the W test is the optimal choice. In contrast to previous studies, our method is general and does not rely on the strict assumption that a common factor causes the cross-sectional dependency. In an empirical application, our method is also demonstrated in terms of the Fisher effect—a hypothesis about the existence of which there is still no clear consensus. Based on our sample of the five Nordic countries we utilize our powerful test and discover evidence which, in contrast to most previous research, confirms the Fisher effect.     

Testing autocorrelation in a system perspective testing autocorrelation

The Breusch-Godfrey test for autocorrelated errors is generalised to cover systems of equations, and the properties of 18 versions of the test are studied using Monte Carlo methods. We show that only one group of tests regularly has actual size close to the nominal size; namely the likelihood ratio tests of the auxiliary regression system that are corrected in some manner for degrees-of-freedom. The Rao Ftest exhibits the best performance, whilst the commonly used TR2 test behaves badly even in single equations. However, the size and power properties of all tests deteriorate sharply as the number of equations increases, the system becomes more dynamic, the exogenous variables become more autocorrelated and the sample size decreases. This performance has, in general, an unknown degree since the interaction amongst these factors does not permit a predictive summary, as might be hoped for by response surface-type approaches.