ERIC Number: ED173207
Record Type: RIE
Publication Date: 1978-Aug
Final Report for Dynamic Models for Causal Analysis of Panel Data. Models for Change in Quantitative Variables, Part III: Estimation from Panel Data. Part II, Chapter 5.
Hannan, Michael T.
This document is part of a series of chapters described in SO 011 759. Addressing the problems of studying change and the change process, the report argues that sociologists should study coupled changes in qualitative and quantitative outcomes (e.g., marital status and earnings). The author presents a model for sociological studies of change in metric variables and focuses on the practical details of using stochastic differential equations (SDEs) in panel analysis. Eleven sections comprise the report. The first considers two broad approaches to empirical studies of SDEs: estimating structural parameters directly from integral equations or using discrete approximations. After arguing for the first approach, the author addresses issues such as autocorrelation of disturbances, unit-specific effects, pooled cross-section and time series estimators, fixed effect estimators, random effects estimators, Monte Carlo studies of small sample properties, and unequally spaced observations. A continuous-time perspective for dealing with problems of analyzing unequally spaced panel data is proposed. The report concludes that researchers can make use of available methods to solve many of the practical problems that arise in applying continuous-time, continuous-state models in sociological research. (Author/KC)
Publication Type: Reports - Research
Education Level: N/A
Sponsor: National Inst. of Education (DHEW), Washington, DC.; National Science Foundation, Washington, DC.
Authoring Institution: Center for Advanced Study in the Behavioral Sciences, Stanford, CA.; Stanford Univ., CA.
Note: For related documents, see SO 011 759-772