作者: William Nelson Goetzmann , None
DOI: 10.1007/BF00153997
关键词: Multicollinearity 、 Economics 、 Estimator 、 Real estate 、 Statistics 、 Regression 、 Generalized least squares 、 Econometrics 、 Bayesian probability 、 M-estimator 、 Estimation
摘要: Simulation techniques allow us to examine the behavior and accuracy of several repeat sales regression estimators used construct real estate return indices. We show that generalized least squares (GLS) method is maximum likelihood estimator, we how estimation can be significantly improved through a Baysian approach. In addition, introduce biased procedure based upon James Stein address problems multicollinearity common procedure.