Alexander Basilevsky

Missing data estimators in the general linear model: an evaluation of simulated data as an experimental design

Previous simulations have reported second order missing data estimators to be superior to the more straightforward first order procedures such as mean value replacement. These simulations however were based on deterministic comparisonsbetween regression criteria even though simulated sampling is a random procedure. In this paper a simulation structured asan experimental design allows statistical testing of the various missing data estimators for the various regression criteria as well as different regression specifications. Our results indicate that although no missing data estimator is globally best many of the computationally simpler first order methods perform as well as the more expensive higher order estimators, contrary to some previous findings.

Karhunen-Loève Analysis of Historical Time Series with an Application to Plantation Births in Jamaica

Abstract A comparison of two spectral analysis procedures is presented: the frequency domain (Fourier transform) model and the Karhunen-Loeve time domain model. Both models are used in turn to analyze a plantation births series during the period 1880–1938. The principal components model can be adapted as the discrete analogue of the Karhunen-Loeve stochastic integral equation in order to decompose a single time series into trend, cycle, and seasonality. The results indicate that the Karhunen-Loeve decomposition, which has been less popular in applied work than the frequency domain model, can provide the social historian with useful results that are easier to interpret.

Socioeconomic classification of countries: A maximum likelihood factor analysis technique

Abstract This paper presents an alternative method to classify countries on the basis of preselected socioeconomic variables. It utilizes the maximum likelihood factor analysis model (MLFA) which is deemed to be superior to other techniques utilized so far. Data on 21 variables and 64 countries provide a consistent and meaningful classification of countries. The advantages of MLFA over other popular techniques is also discussed.