Showing papers by "Manuel Arellano published in 1999"

Symmetrically normalized instrumental-variable estimation using panel data

Abstract: We discuss the estimation of linear panel-data models with sequential moment restrictions using symmetrically normalized generalized method of moments (GMM) estimators (SNM) and limited information maximum likelihood (LIML) analogues. These estimators are asymptotically equivalent to standard GMM but are invariant to normalization and tend to have a smaller finite-sample bias, especially when the instruments are poor. We study their properties in relation to ordinary GMM and minimum distance estimators for AR(1) models with individual effects by mean of simulations. Finally, as empirical illustrations, we estimate by SNM and LIML employment and wage equations using panels of U.K. and Spanish firms.

The Time Series and Cross-Section Asymptotics of Dynamic Panel Data Estimators

Abstract: In this paper we derive the asymptotic properties of within groups (WG), GMM and LIML estimators for an autoregressive model with random effects when both T and N tend to infinity. GMM and LIML are consistent and asymptotically equivalent to the WG estimator. When T/N->0 the fixed T results for GMM and LIML remain valid, but WG although consistent has an asymptotic bias in its asymptotic distribution. When T/N tends to a positive constant, the WG, GMM and LIML estimators exhibit negative asymptotic biases of order T,N and (2N-T), respectively. In addition, the crude GMM estimator that neglects the autocorrelation in first differenced errors is inconsistent as T/N->c>0, despite being consistent for fixed T. Finally, we discuss the properties of a random effects MLE with unrestricted initial conditions when both T and N tend to infinity.

Analysis of panels and limited dependent variable models: Autoregressive models with sample selectivity for panel data

Abstract: Introduction Recent studies have developed econometric procedures for the analysis of the time series properties of panel data sets consisting of large numbers of short individual time series (e.g., Anderson and Hsiao (1981), Chamberlain (1984), Holtz-Eakin, Newey, and Rosen (1988), and Arellano and Bond (1991)). The analysis is typically based on empirical autoregressive equations including time and individual effects, and possibly observed time-varying exogenous variables. Individual effects are removed by differencing and lagged variables are used as instruments in order to retrieve consistent estimators of the autoregressive coefficients of the levels equation. Alternatively, one could choose moving average processes and components of variance to model the autocovariance matrix of the data in first differences, using methods of moments estimation and testing as well (as done, for example, by Abowd and Card (1989)). In either case, the motivation for this type of analysis with micro data is often to establish a mapping between the observed dynamic interactions and those implied by a theoretical model, or at least to test particular time series implications of such model. The purpose of this chapter is to formulate procedures for the analysis of the time series behavior of panel data subject to censoring. We apply these methods to analyze the dynamics of female labor supply and wages using PSID data. We follow the standard latent variable approach to models with selectivity and assume a linear autoregressive model for a latent variable which is only partly observed due to a selection mechanism.

Learning about Migration Decisions from the Migrants. An Exercise in Endogenous Sampling and Complementary Datasets

Abstract: The authors investigate the determinants of the remarkable increase in intra-regional migrations since the 1980's in Spain, using a large administrative micro dataset on migrants. Conditional migration probabilities are identified by comparing the migrant's joint distribution of characteristics to the corresponding distribution from the Spanish Labour Force Survey. The proportion of employment in the service industry, unemployment, house prices and education, all have important positive effect on the individual probabilities of intra-regional migration.

Learning About Migration Decisions From the Migrants: An Exercise in Endogenous Sampling and Complementary Datasets ∗

Abstract: We investigate the determinants of the remarkable increase in intra-regional migrations since the 1980’s in Spain, using a large administrative micro dataset on migrants. Conditional migration probabilities are identified by comparing the migrants’ joint distribution of characteristics to the corresponding distribution from the Spanish Labour Force Survey. The proportion of employment in the service industry, unemployment, house prices and education, all have an important positive effect on the individual probabilities of intra-regional migration.