SMALL-SAMPLE PROPERTIES OF VARIOUS SIMULTANEOUS-EQUATION ESTIMATORS: THE RESULTS OF SOME MONTE-CARLO EXPERIMENTS,
Abstract
This study reports the results of 172 Monte Carlo experiments on the small-sample properties of some simultaneous equation estimators. The estimators studied were Direct Least-Squares, Two-Stage Least-Squares, Nagar's Unbiased k-Class Estimator, Limited-Information Maximum-Likelihood, Three-Stage Least-Squares, FullInformation Maximum-Likelihood and Least-Squares used to estimate the reduced form. The estimates of the structural equations, their standard errors, the estimates of the variance-covariance matrix of disturbances, the estimates of the reduced-form coefficients and the predictive ability of the methods were examined. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 01, 1964
- Accession Number
- AD0610085
Entities
People
- J. G. Cragg
Organizations
- Princeton University