Forum: open-discussion

RE: Significance of beta coeffients of the restricted stochastic frontier model [ Reply ]
By: Arne Henningsen on 2013-03-13 09:23

[forum:39167]

Dear Luke

Please read the JPA paper that you mentioned. It says:

The restricted parameters (b^0) are asymptotically equivalent to a (successful) restricted onestep ML estimation (Koebel et al. 2003). However, it might be problematic to obtain a (consistent) covariance matrix of the restricted parameters sigmaHat^0_b, because standard bootstrapping leads to an inconsistent covariance matrix if the restricted parameters are at the boundary of the feasible parameter space (Andrews 2000; Dhrymes 2006). Andrews (2000) suggests alternative methods, e.g. rescaled bootstrapping, that lead to a consistent covariance matrix even in the case of binding inequality constraints. However, these alternative methods are only valid under speciﬁc conditions that need to be checked for our speciﬁc case. Thus we leave this interesting topic for future research.

Best regards,
Arne

Significance of beta coeffients of the restricted stochastic frontier model [ Reply ]
By: Luke Abatania on 2013-03-13 06:20

[forum:39166]

I have used the R code available in the paper "Imposing regional monotonicity on translog production frontiers with a simple three-step procedure" by Arne Hennigsen, Christian H. C.A. Henning to estimate a stochastic frontier model. How can I determine the significance of the beta coefficients of the model as well as the significance of the adjusted coefficients. Any ideas on computing either the standard errors, t-values or p-values for these parameter estimates would be useful.