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.

We derive a neat and compact representation of the asymptotic Fisher information matrix of a vector ARMA process. Its inverse can be used immediately as the asymptotic covariance matrix of the Gaussian maximum likelihood estimator. We also provide the robust sandwich covariance estimator when...

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.

In this paper we provide a general solution to the problem of controlling the probability of a type I error in normality tests for the disturbances in linear regressions when using robust-regression residuals. We show that many classes of well-known robust regression estimators belong to the...

This paper investigates the test of joint significance for binary choice model with multiple integrated explanatory variables. It is found that for the widely used logit and probit models, even though the estimators have a different convergence rate under null hypothesis compared with the case...

In this note, we argue that tests of overidentifying restrictions give little information on the validity of the moment conditions implied by the underlying economic model, and therefore are mute about the possibility of identifying the parameters of interest.

Standard tests are generally not applicable in panel data models with selection. The paper shows how the Hausman specification test and the Sargan–Hansen test for overidentifying restrictions can be generalized to panel data models with unobserved heterogeneity and sample selection.

A Wald type test of the joint null hypothesis of linearity and nonstationarity within a threshold autoregressive process of order one with deterministic components is developed. Its limiting distribution is derived and its local power and finite sample properties investigated.

We suggest a new unit-root test with a Fourier function in the deterministic term in a Dickey–Fuller type regression framework. Our suggested test can complement the Fourier LM and DF-GLS unit root tests. They have good size and power properties.