Search results

Standard multivariate tests of mean variance efficiency (MVE) have been criticised on the grounds that they require regression residuals to have a multivariate normal distribution. Generally, the existing evidence suggests that the normality assumption is questionable, even for monthly returns. MacKinlay and Richardson (1991) developed a generalised method of moments (GMM) framework which provides tests which are valid under much weaker distributional assumptions. They examined monthly US data formed into size based portfolios, for mean‐variance efficiency relative to the Sharpe‐Lintner CAPM. They found that inferences regarding mean‐variance efficiency can be sensitive to the test considered. In this paper we further investigate their GMM tests using monthly Australian data over the period 1974 to 1994. We extend upon their analysis to consider an alternative version of their GMM test and also to examine a zero‐beta version of the CAPM. Similar to the US case, our results also indicate sensitivity of inferences to the tests used. Finally, while we find that the GMM tests generally provide rejection of mean‐variance efficiency, tests involving the zero‐beta CAPM, particularly when a value‐weighted market index is used, prove less prone to rejection.

Goodness‐of‐fit test based on K_ϕ‐divergence between observed and theoretical frequencies are considered. The asymptotic chi‐square null distribution and three alternative approximations to the exact distribution function of this family are compared in small samples. Numerical results are presented for the symmetric null hypothesis for different multinomial sample sizes with various cell numbers. Exact power under specific alternatives to the symmetric null hypothesis are calculated and a comparison with the family of power divergence statistics is made.

Measurement of diminishing or divergent cross section dispersion in a panel plays an important role in the assessment of convergence or divergence over time in key economic indicators. Econometric methods, known as weak σ-convergence tests, have recently been developed (Kong, Phillips, & Sul, 2019) to evaluate such trends in dispersion in panel data using simple linear trend regressions. To achieve generality in applications, these tests rely on heteroskedastic and autocorrelation consistent (HAC) variance estimates. The present chapter examines the behavior of these convergence tests when heteroskedastic and autocorrelation robust (HAR) variance estimates using fixed-b methods are employed instead of HAC estimates. Asymptotic theory for both HAC and HAR convergence tests is derived and numerical simulations are used to assess performance in null (no convergence) and alternative (convergence) cases. While the use of HAR statistics tends to reduce size distortion, as has been found in earlier analytic and numerical research, use of HAR estimates in nonparametric standardization leads to significant power differences asymptotically, which are reflected in finite sample performance in numerical exercises. The explanation is that weak σ-convergence tests rely on intentionally misspecified linear trend regression formulations of unknown trend decay functions that model convergence behavior rather than regressions with correctly specified trend decay functions. Some new results on the use of HAR inference with trending regressors are derived and an empirical application to assess diminishing variation in US State unemployment rates is included.

We extend Vuong’s (1989) model-selection statistic to allow for complex survey samples. As a further extension, we use an M-estimation setting so that the tests apply to general estimation problems – such as linear and nonlinear least squares, Poisson regression and fractional response models, to name just a few – and not only to maximum likelihood settings. With stratified sampling, we show how the difference in objective functions should be weighted in order to obtain a suitable test statistic. Interestingly, the weights are needed in computing the model-selection statistic even in cases where stratification is appropriately exogenous, in which case the usual unweighted estimators for the parameters are consistent. With cluster samples and panel data, we show how to combine the weighted objective function with a cluster-robust variance estimator in order to expand the scope of the model-selection tests. A small simulation study shows that the weighted test is promising.

We analyze Lagrange Multiplier (LM) tests for a shift in trend of a univariate time series at an unknown date. We focus on the class of LM statistics based on nonparametric kernel estimates of the long run variance. Extending earlier work for models with nontrending data, we develop a fixed-b asymptotic theory for the statistics. The fixed-b theory suggests that, for a given statistic, kernel, and significance level, there usually exists a bandwidth such that the fixed-b asymptotic critical value is the same for both I(0) and I(1) errors. These “robust” bandwidths are calculated using simulation methods for a selection of well-known kernels. We find when the robust bandwidth is used, the supremum statistic configured with either the Bartlett or Daniell kernel gives LM tests with good power. When testing for a slope change, we obtain the surprising finding that less trimming of potential shift dates leads to higher power, which contrasts the usual relationship between trimming and power. Finite sample simulations indicate that the robust LM statistics have stable size with good power.