We propose an empirical likelihood-based method of inference for comparing inequality between two populations. A series of Monte Carlo experiments are used to assess our method’s finite sample performance. We illustrate our approach using some Canadian household income data.

Using an empirical likelihood approach, we show that generalized linear models can still be consistently estimated even if dependent variables are not missing at random, and derive a Hausman test by comparing this estimator to the standard one.

A new method of assessing the comparative quality of forecasting models is introduced. This method focuses on the quality of forecasting models over a set of series (cf. the traditionally adopted series-by-series approach)–with a forecasting model that produces good forecasts over a series set...

We propose a test of bivariate stochastic dominance within a generalized framework for testing inequality restrictions, utilizing the covariance structure of the estimates of the joint distribution functions. Monte Carlo simulations and an empirical example assess its usefulness.

This paper proposes a cumulative sum (CUSUM) based statistic to test if there is a common variance change-point in panel data models. Asymptotic distribution is derived under the null hypothesis and the consistency of the test is proven under the alternative hypothesis. Monte Carlo experiment is...

This paper develops new model selection criteria for regression with heteroskedastic and autocorrelated errors. We prove the selection consistency of the introduced criteria and evaluate their performance by simulation. The results suggest that the new criteria may bring significant improvement...

The commonly-used version of the double-hurdle model rests on a rather restrictive set of statistical assumptions, which are very seldom tested by practitioners, mainly because of the lack of a standard procedure for doing so, although violation of such assumptions can lead to serious modelling...

The use of recursive demeaning and detrending procedures in unit root tests has been popular in the literature, since they lead to more precise estimation of the persistence parameter and greater power in unit root tests. However, we find that unit root tests using these recursive procedures...

This paper proposes a test for linearity against exponential smooth transition models with endogenous right-hand-side variables: to the very best of our knowledge, this class of models is new to the literature. By Monte Carlo analysis the test is shown to have good finite sample properties.

We examine the performance of nonlinear instrumental variable (NIV) unit root tests using various recursive detrending methods. We find that the NIV unit root tests using the recursive detrending method of Chang (2002) are the most powerful. They are more powerful than OLS based DF tests.