From S.U-Din at massey.ac.nz Tue Sep 12 10:18:10 2017 From: S.U-Din at massey.ac.nz (U-Din, Salah) Date: Tue, 12 Sep 2017 00:18:10 +0000 Subject: [Limdep Nlogit List] SFA Message-ID: Hi I am working on Stochastic Frontier Analysis under Battese and Coelli (1995) model for panel data. I didn't find the commands for BC95 in panel data. Given manual has only for BC92 in panel data and BC95 for time series only. Please can you send me some material which has discussed commands for BC95 for panel data. Thanks Salah 0902-AU-NS-01373 From wgreene at stern.nyu.edu Tue Sep 12 20:49:09 2017 From: wgreene at stern.nyu.edu (William Greene) Date: Tue, 12 Sep 2017 06:49:09 -0400 Subject: [Limdep Nlogit List] SFA In-Reply-To: References: Message-ID: Salah. It is in section E63.3.2 Battese and Coelli (1995) Formulation in the manual, with an example. Please refer to the original article for descriptive material. /B. Greene On Mon, Sep 11, 2017 at 8:18 PM, U-Din, Salah via Limdep < limdep at mailman.sydney.edu.au> wrote: > Hi > I am working on Stochastic Frontier Analysis under Battese and Coelli > (1995) model for panel data. I didn't find the commands for BC95 in panel > data. Given manual has only for BC92 in panel data and BC95 for time series > only. Please can you send me some material which has discussed commands for > BC95 for panel data. > > Thanks > Salah > 0902-AU-NS-01373 > > _______________________________________________ > Limdep site list > Limdep at mailman.sydney.edu.au > http://limdep.itls.usyd.edu.au > > -- William Greene Department of Economics Stern School of Business, New York University 44 West 4 St., 7-90 New York, NY, 10012 URL: https://protect-au.mimecast.com/s/K410BnUG9w7lIR?domain=people.stern.nyu.edu Email: wgreene at stern.nyu.edu Ph. +1.212.998.0876 Editor in Chief: Journal of Productivity Analysis Editor in Chief: Foundations and Trends in Econometrics Associate Editor: Economics Letters Associate Editor: Journal of Business and Economic Statistics Associate Editor: Journal of Choice Modeling From S.U-Din at massey.ac.nz Fri Sep 15 12:25:56 2017 From: S.U-Din at massey.ac.nz (U-Din, Salah) Date: Fri, 15 Sep 2017 02:25:56 +0000 Subject: [Limdep Nlogit List] SFA In-Reply-To: References: Message-ID: Hi William I am looking for one-stage time varying SFA. First I tried BC92 (E64.6) with following codes: Frontier;Lhs=tc;Cost;Rhs=one,w1,w2,w11,w12,w22,y1,y2,y3,y11,y22,y33,y12,y13,y23,w1y1,w1y2,w1y3,w2y1,w2y2,w2y3,tr,tsq;model=bc;costeff=ce;hfu=one,z1,z2,z3,z4,z5,z6,z7,z8,z9;panel$ Where Tc(Total Cost) is dependent variable, w1,w2 are inputs, y1,y2,y3 are outputs of bank. These codes give me time varying cost efficiency but without any influence of z variable in hfu. Also didn't give relationship of z variables with inefficiency. Then I tried as per your recommendation the BC95 (E63.3.2 & E63.3.4) codes as: Frontier;Lhs=tc;Cost;Rhs=one,w1,w2,w11,w12,w22,y1,y2,y3,y11,y22,y33,y12,y13,y23,w1y1,w1y2,w1y3,w2y1,w2y2,w2y3,tr,tsq;model=bc95;costeff=ce;Rh2=one,z1,z2,z3,z4,z5,z6,z7,z8,z9 ;panel$ It gives me following error: Iterative procedure has converged Normal exit: 31 iterations. Status=0, F= -.1569414D+03 Error 143: Models - estimated variance matrix of estimates is singular Same error does came after the running the BC92 codes mentioned above but it produced results but BC95 codes didn't produced any results. Please can you advice what is wrong with my codes? I want time varying cost efficiency under influence of z(exogenous) variables and their relationship with inefficiency or efficiency. Anyone can advise please if know any possible solution. Thanks Salah -----Original Message----- From: Limdep [mailto:limdep-bounces at mailman.sydney.edu.au] On Behalf Of William Greene Sent: Tuesday, September 12, 2017 10:49 PM To: Limdep and Nlogit Mailing List Subject: Re: [Limdep Nlogit List] SFA Salah. It is in section E63.3.2 Battese and Coelli (1995) Formulation in the manual, with an example. Please refer to the original article for descriptive material. /B. Greene On Mon, Sep 11, 2017 at 8:18 PM, U-Din, Salah via Limdep < limdep at mailman.sydney.edu.au> wrote: > Hi > I am working on Stochastic Frontier Analysis under Battese and Coelli > (1995) model for panel data. I didn't find the commands for BC95 in > panel data. Given manual has only for BC92 in panel data and BC95 for > time series only. Please can you send me some material which has > discussed commands for > BC95 for panel data. > > Thanks > Salah > 0902-AU-NS-01373 > > _______________________________________________ > Limdep site list > Limdep at mailman.sydney.edu.au > http://limdep.itls.usyd.edu.au > > -- William Greene Department of Economics Stern School of Business, New York University 44 West 4 St., 7-90 New York, NY, 10012 URL: https://protect-au.mimecast.com/s/lqmYBbSDmWo3tm?domain=people.stern.nyu.edu Email: wgreene at stern.nyu.edu Ph. +1.212.998.0876 Editor in Chief: Journal of Productivity Analysis Editor in Chief: Foundations and Trends in Econometrics Associate Editor: Economics Letters Associate Editor: Journal of Business and Economic Statistics Associate Editor: Journal of Choice Modeling _______________________________________________ Limdep site list Limdep at mailman.sydney.edu.au http://limdep.itls.usyd.edu.au From lars.persson at umu.se Thu Sep 28 23:06:10 2017 From: lars.persson at umu.se (Lars Persson) Date: Thu, 28 Sep 2017 13:06:10 +0000 Subject: [Limdep Nlogit List] Test for differences in marginal wtp between samples Message-ID: <9D39AD91-CA2D-4120-8DD4-AA1A03C33B0C@umu.se> Dear all I have a split sample choice experiment and would like to test if there is a significant difference in marginal wtp between the samples. So far, the model specification is RPL in preference space with normal distributions except for the cost attribute (fixed). I have seen people doing the complete combinatorial test suggested by Poe et al. 2005, is this still the recommendation? If so, I would really need and appreciate practical guidance on how to do this. Related to this issue, I have calculated the marginal wtp?s as the ratio of coefficients using the Wald command. I know it is recommended to use the ;wtp command to get the individual wtp?s, but how would that affect the use of the ?Poe test?? Any alternatives? Any advise or help related to these topics are deeply appreciated. Lars