More documents from Piatek, Rémi

Export to your Reference Manger

Please Copy & Paste



Bookmark and Share

Bayesian inference for factor structure models via gibbs sampling

[working paper]

Piatek, Rémi

fulltextDownloadDownload full text

(374 KByte)

Citation Suggestion

Please use the following Persistent Identifier (PID) to cite this document:

Further Details
Corporate Editor Universität Konstanz, Center for Quantitative Methods and Survey Research (CMS)
Abstract "The goal of this paper is to provide all the technical details required to implement Gibbs sampling for the estimation of simultaneous equation models with common latent factors among their regressors, so-called "factor structure models". Linear, dichotomous and censored response models, as well as ordered and unordered response models can be accommodated in this framework. The latent factors can be either correlated or not, and specified as normally distributed or as following a finite mixture of normal distributions for more flexibility. All conditional distributions are derived and can be used to construct the Gibbs sampler step by step." (author's abstract)
Keywords factor analysis; model analysis; social science; economics; statistical method; indicator; estimation
Classification Sociology of Economics
Document language English
Publication Year 2010
City Konstanz
Page/Pages 26 p.
Series CMS Discussion Paper, 3
Status Published Version
Licence Deposit Licence - No Redistribution, No Modifications
data provider This metadata entry was indexed by the Special Subject Collection Social Sciences, USB Cologne