http://peacelovekenzie.life-and-things.com/ Ñ�аÑ�Ñ�каз и Ñ�зÑ�к zh-tw http://peacelovekenzie.life-and-things.com/2008/09/15/ols-and-probit-diff.html n<-10000 disper<-0.6 x<-rnorm(n) e<-rnorm(n,disper) a<-0.3;b<-0.6; y<-a+b*x+e y<-ifelse(y<0,0,1) g<-y~x summary(lm(g)) summary(glm(g,family=binomial(link=probit))) Title: Results on the bias as well as inconsistency of ordinary least squares on behalf of the linear probability model Author William C. Horracea, and Ronald L. Oaxacab Limitations of the Linear Probability Model (LPM) are well-known. OLS estimated probabilities are not bounded on the unit interval, as well as OLS estimation implies that heteroscedasticity exists. Conventional advice points to probit or logit as the standard remedy, which bound the maximum likelihood estimated probabilities on the unit interval. However, the fact that consistent estimation of the LPM may be difficult does not imply that either probit or logit is the correct specification of the probability model; it may be reasonable to assume that probabilities are generated from bounded linear decision rules. Theoretical rationalizations on behalf of the LPM are in Rosenthal (1989) as well as Heckman as well as Snyder, 1977 J.J. Heckman as well as J.M. Snyder Jr., Linear probability models of the demand on behalf of attributes with an empirical application to estimating the preferences of legislators, Rand Journal of Economics 28 (1977), pp. S142–S189.Heckman as well as Snyder (1977). Despite the attractiveness of logit as well as probit on behalf of estimating binary dependent variable models, OLS on the LPM is still used. Recent applications include Klaassen as well as Magnus (2001), Bettis as well as Fairlie (2001), Lukashin (2000), McGarry (2000), Fairlie as well as Sundstrom (1999), Reiley (2005), as well as Currie as well as Gruber (1996). Empirical rationales on behalf of the LPM specification are plentiful. McGarry appeals to ease of interpretation of estimated marginal effects, while Reiley cites a perfect correlation problem associated with the probit model. Fairlie as well as Sundstrom prefer LPM because of the fact that it implies a simple expression on behalf of the modification in unemployment rate between two censuses. Bettis as well as Farlie opt on behalf of LPM because of the fact that of an extremely large sample size as well as other simplifications implied by it. Lukashin uses the LPM, because of the fact that it lends itself to a model selection algorithm based on an adaptive gradient criterion. Currie as well as Gruber state that logit, probit, as well as OLS are similar on behalf of their data as well as only report LPM results. Other rationales on behalf of the OLS on the LPM are complications of probit/logit models in certain contexts. Klaassen as well as Magnus cite panel data complications in their tennis example as well as select OLS. OLS is perhaps justified in simultaneous equations/instrumental variable methods. The presence of dummy endogenous regressors is problematic if the DGP is assumed to be probit or logit; these difficulties were first considered by Heckman (1978 ). While perhaps less popular than logit as well as probit, OLS on the LPM model still finds its way into the literature on behalf of various reasons. Some well-known LPM theorems are provided in Amemiya (1977). Econometrics textbooks (e.g., Greene, 2000), acknowledge complications leading to biased as well as inconsistent OLS estimates. Nevertheless, the literature is not clear on the precise conditions when OLS is problematic. This note rigorously lays out these conditions, derives the finite-sample as well as asymptotic biases of OLS, as well as provides additional results that highlight the appropriateness or inappropriateness of OLS estimation of the LPM. Finally, we suggest a trimmed sample estimator that could reduce OLS bias. Mon, 15 Sep 2008 19:35:07 -0400 peacelovekenzie