We propose an Instrumental Variable method for Generalised Accelerated Failure Time (GAFT) models that adjust for possible endogeneity of the intervention of interest, without suffering the problems of the intention-to-treat method. We develop an estimatiom procedure that collapses to the linear...

In view of higher fertility and mortality rates in Pakistan compared to India, this paper examines the two-way relationship between birth interval and child mortality and compares the behaviour of households in the Indian and Pakistani provinces of Punjab. Birth interval and child survival are...

This paper examines the two-way relationship between birth interval and child survival and compares the behaviour of households in the Indian and Pakistani provinces of Punjab. Birth interval and child survival are modelled here as correlated hazard processes, allowing for mother- specific...

We present evidence on the duration of firm-bank relationships using a unique panel data set of connections between Oslo Stock Exchange-listed firms and their banks for the period 1979-1994. We focus on the determinants of the duration of a relationship and the causes for ending an existing bank...

Classical parametric estimation methods applied to nonlinear regression and limited-dependent-variable models are very sensitive to misspecification and data errors. This sensitivity is addressed by the theory of robust statistics which builds upon parametric specification, but provides...

Distributions for returns are used to compute the capital charge for portfolios in investment banks. The mainstream definition of returns is based on closing prices and neglects the important effects of intraday trading activity on the losses . In this paper we introduce ''minimal returns'', a...

The paper seeks to fill the void in the empirical economics literature of U.S. antimerger policy. A new empirical measure of relief secured in a Section 7 case is proposed. The extent of obtained relief is modeled as a fraction of the competitive overlap subject to structural divestiture. The...

Since recent immigrants tend to earn less than natives, their relative labor market status has been adversely impacted by an increase in the return to labor market skills and widening wage inequality over the past two decades. To evaluate the magnitude of this effect, this study uses Social...

This study uses Social Security earnings records matched to recent cross-sections of the SIPP and CPS to study the earnings progress of U.S. immigrants.The data show that immigrants' earnings grow 10 to 13 percent during their first twenty years in the U.S. relative to the earnings of natives...