Two model averaging approaches are used and compared in estimating and forecasting dynamic factor models, the well-known BMA and the recently developed WALS. Both methods propose to combine frequentist estimators using Bayesian weights. We apply our framework to the Armenian economy using...

This paper presents some two-step estimators for a wide range of parametric panel data models with censored endogenous variables and sample selection bias. Our approach is to derive estimates of the unobserved heterogeneity responsible for the endogeneity/selection bias to include as additional...

This paper uses longitudinal data to perform tests of asymmetric information in the French automobile insurance market for the 1995-1997 period.This market is characterized by the presence of a regulated experience-rating scheme (bonusmalus).We demonstrate that the result of the test depends...

The panel-data regression models are frequently applied to micro-level data, which often suffer from data contamination, erroneous observations, or unobserved heterogeneity. Despite the adverse effects of outliers on classical estimation methods, there are only a few robust estimation methods...

We examine the importance of possible non-random attrition to an econometric model of life cycle labor supply including joint nonlinear taxation of wage and interest incomes and latent heterogeneity.We use a Wald test comparing attriters to nonattriters and variable addition testing based on...

Using panel data of 58 developing countries for the period 1980-1998, this study shows that the responsiveness of the $2 a day poverty headcount measure to changes in mean income and inequality significantly decreases with initial inequality and the ratio poverty line over mean income - taken as...

In many industrial countries, ownership rates of risky assets have risen substantially over the past decade.This trend has potentially wide-ranging implications for the intertemporal and cross-sectional allocation of risk, and for the macro economy, establishing the need for understanding...

We propose a method to forecast the winner of a tennis match, not only at the beginning of the match, but also (and in particular) during the match.The method is based on a fast and exible computer program TENNISPROB, and on a statistical analysis of a large data set from Wimbledon, both at...