From medakseth at gmail.com Wed Aug 1 20:13:04 2018 From: medakseth at gmail.com (medard kakuru) Date: Wed, 1 Aug 2018 13:13:04 +0300 Subject: [Limdep Nlogit List] Inquiry Message-ID: Dear Forum members, Hope you are well. Kindly help me out with this. I have estimated three models: mixed logit (M1), GMXL in utility space (M2) and GMXL in wtp space (M3). I expected the estimates of M3 to be different from the other two models because the estimates of M3 are WTP estimates. for my case all the parameter estimates in the three models are similar (almost). See the commands and the output for the three models in the attachment. Kindly advise/comment. Thanks, Medard From ambi.ganesh at gmail.com Fri Aug 3 16:09:54 2018 From: ambi.ganesh at gmail.com (Ganesh A R) Date: Fri, 3 Aug 2018 11:39:54 +0530 Subject: [Limdep Nlogit List] Regarding calculating probabilities from ordered logit models Message-ID: Dear Sir, I am trying to manually calculate the probabilities from the estimated coefficients of an ordered logit model. I am currently using ordinal data with responses 0,1,2 and 3. The LIMDEP manual says that if a constant "ONE" is included in the specification, then the first threshold is considered to be zero. Hence using that when I manually calculated the probability of each response, it doesn't match with the probabilities that are calculated via the model itself using the " ; prob = name $" function. Following is the method I used for calculating probabilities. U = bo + b1x1 + b2x2 + ..... + bkxk Where, U = utility function [b] = coefficients estimated from the model [x] = explanatory variables f(0) = 1 / (1 + exp(-U)) f(1) = 1 / (1 + exp(mu(1) - U) f(2) = 1 / (1 + exp(mu(2) - U) prob(0) = f(0) prob(1) = f(1) - f(0) prob(2) = f(2) - f(1) prob(3) = 1 - f(2) where, mu(1) and mu(2) are the threshold parameters obtained from the model f(0), f(1) and f(2) is the CDF prob(k) = probability of response k Thanking you Yours Sincerely Ganesh Ambi Ramakrishnan Doctoral Student From wgreene at stern.nyu.edu Wed Aug 8 20:41:12 2018 From: wgreene at stern.nyu.edu (William Greene) Date: Wed, 8 Aug 2018 06:41:12 -0400 Subject: [Limdep Nlogit List] Regarding calculating probabilities from ordered logit models In-Reply-To: References: Message-ID: Dear Ganesh Ambi Ramakrishnan You are using the wrong formulas for f0, f1 and f2. The sample program below illustrates. Regards Bill Greene sample ; 1 - 1000 $ calc ; ran(12345)$ create ; x1=rnn(0,1) ; x2 = rnn(0,1) ; u = rnu(0,1) ; e = log(u/(1-u))$ create ; u = x1 + x2 + e $ calc ; mu1 = 1 ; mu2 = 2 $ create ; y = 0 $ create ; if (u <= 0) y = 0 $ creare ; if (u > 0 & u <= 1)y = 1$ create ; if (u > 1 & u <= 2)y = 2 $ creare ; if (u > 2) y = 3 $ histogram ;rhs=y$ ordered ; logit ; lhs = y ; rhs = one,x1,x2 ; prob=pfit $ calc ; mu1 = mu(1) ; mu2 = mu(2) $ create ; ufit = b(1)+b(2)*x1+b(3)*x2$ create ; f0=exp(-ufit)/(1+exp(-ufit))$ create ; f1=exp(mu1-ufit)/(1+exp(mu1-ufit))$ create ; f2=exp(mu2-ufit)/(1+exp(mu2-ufit))$ create ; p0=f0 ; p1=f1-f0 ; p2=f2-f1 ; p3=1-f2$ create ; pmodel=(y=0)*p0 + (y=1)*p1 + (y=2)*p2 + (y=3)*p3$ dsta;rhs=p0,p1,p2,p3,pfit,pmodel$ On Fri, Aug 3, 2018 at 2:09 AM, Ganesh A R via Limdep < limdep at mailman.sydney.edu.au> wrote: > Dear Sir, > > I am trying to manually calculate the probabilities from the estimated > coefficients of an ordered logit model. > I am currently using ordinal data with responses 0,1,2 and 3. > The LIMDEP manual says that if a constant "ONE" is included in the > specification, then the first threshold is considered to be zero. Hence > using that when I manually calculated the probability of each response, it > doesn't match with the probabilities that are calculated via the model > itself using the " ; prob = name $" function. > > Following is the method I used for calculating probabilities. > > U = bo + b1x1 + b2x2 + ..... + bkxk > Where, > U = utility function > [b] = coefficients estimated from the model > [x] = explanatory variables > > f(0) = 1 / (1 + exp(-U)) > f(1) = 1 / (1 + exp(mu(1) - U) > f(2) = 1 / (1 + exp(mu(2) - U) > prob(0) = f(0) > prob(1) = f(1) - f(0) > prob(2) = f(2) - f(1) > prob(3) = 1 - f(2) > where, > mu(1) and mu(2) are the threshold parameters obtained from the model > f(0), f(1) and f(2) is the CDF > prob(k) = probability of response k > > Thanking you > > Yours Sincerely > > Ganesh Ambi Ramakrishnan > > Doctoral Student > _______________________________________________ > 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/ewMVCr8DLRtoOmXXC7w_Ma?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 p.cyrenne at uwinnipeg.ca Tue Aug 14 23:49:38 2018 From: p.cyrenne at uwinnipeg.ca (Phil Cyrenne) Date: Tue, 14 Aug 2018 13:49:38 +0000 Subject: [Limdep Nlogit List] WTp estimation In-Reply-To: References: <22E7A439-EAF7-43E1-B7ED-AC0731F663EC@economics.gu.se>, Message-ID: <244ecc218f6442699989dbd759b34dbc@uwinnipeg.ca> Hello Professor Greene: I am sure you get swamped with econometric questions, but I have looked through your textbook and have this particular question about estimating a simultaneous equation system. With an endogenous regressor, in general an IV variable is needed, which as you know searching for an appropriate one can prove challenging. I have been under the impression that in the case of an endogenous regressor, the common alternative has been to use the lagged value of the endogenous regressor, since in general it is strongly correlated with the endogenous regresso but is less likely to be correlated with the other endogenous regressor. I have thought this was a relatively uncontroversial assumption. However, I suppose the lagged value of the endogenous regressor can be viewed as an IV variable, but I have not come across papers where the standard IV tests have been applied to the use of a lagged endogenous variable as a particular IV variable. My question, are these tests really needed when a lagged endogenous variable has been used as instrument? If no, why? If it is, Is it a common practice? Finally is there a handy reference for this practice? Thanks in advance. Phil Cyrenne ________________________________ From: Limdep on behalf of William Greene via Limdep Sent: Tuesday, July 24, 2018 9:51:14 AM To: Limdep and Nlogit Mailing List Cc: William Greene Subject: Re: [Limdep Nlogit List] WTp estimation exactly right. I agree Bill Greene On Tue, Jul 24, 2018 at 10:01 AM, Fredrik Carlsson < fredrik.carlsson at economics.gu.se> wrote: > You have one less parameter, so reduce the number of parameters in the > labels part of the Wald-command > > Fredrik > > Skickat fr?n min iPhone > > > 24 juli 2018 kl. 15:53 skrev medard kakuru via Limdep < > limdep at mailman.sydney.edu.au>: > > > > Thanks Bill. > > I am facing another challenge. I use the wald command to estimate wtp > after > > estimating a mixed logit with all non-attribute coefficients given a > normal > > distribution and the syntax works. Since the standard deviation for one > > attribute was not significant, I re-specify the model and make that > > attribute non-random. Running the wald command after this specification > > gives me an error message that "wrong number of start values - need one > for > > each label". What could be the problem? > > > > Thanks. > > > > Medard > > > > On Sun, Jul 22, 2018 at 12:35 AM, William Greene via Limdep < > > limdep at mailman.sydney.edu.au> wrote: > > > >> Medard. WTP for all variants of the RP logit model is described in > >> Chapter N29. That would be good for the S-MNL and GMXL. I'm > >> not sure what the G-MNL model is. WTP is generally a ratio of > >> coefficients, though in RP models, the WTP is computed for each > >> individual. P values for LR statistics are generally not computed. > >> I'm not sure what the P-value at 5% is. You can use CALC to look > >> up critical values for chi-squared if you wish. The function > >> CTB(prob,df) returns the critical value for probability prob with > >> degrees of freedom df. > >> Regards > >> /Bill Greene > >> > >> On Fri, Jul 20, 2018 at 5:22 AM, medard kakuru via Limdep < > >> limdep at mailman.sydney.edu.au> wrote: > >> > >>> Dear All, > >>> Kindly help me with the syntax for wtp in preference space. I am > running > >>> S-MNL, G-MNL and the GMXL models and want to compare WTP estimates for > >> the > >>> three models. I also want to compare these estimates with those > obtained > >>> from wtp-space model estimations. I understand that for a simple MXL > >> model, > >>> the syntax for wtp is just a ratio of attribute and cost coefficients. > Is > >>> it the case with the models I am estimating? I expected to find the > >> syntax > >>> in the Nlogit manual (section N33) but it's not listed. > >>> > >>> > >>> Secondly, I have seen the syntax for carrying out a loglikelihood ratio > >>> test and the syntax gives the LR statistic. How do i go further to get > a > >>> p-value (say at 5%) for the statistic? > >>> > >>> Thanks in advance. > >>> > >>> Regards, Medard > >>> _______________________________________________ > >>> 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/SXT7CQnzP0trQ7YVuxpOHa?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 > >> > >> > > _______________________________________________ > > Limdep site list > > Limdep at mailman.sydney.edu.au > > http://limdep.itls.usyd.edu.au > > > _______________________________________________ > 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/SXT7CQnzP0trQ7YVuxpOHa?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 wgreene at stern.nyu.edu Wed Aug 15 01:15:45 2018 From: wgreene at stern.nyu.edu (William Greene) Date: Tue, 14 Aug 2018 11:15:45 -0400 Subject: [Limdep Nlogit List] WTp estimation In-Reply-To: <244ecc218f6442699989dbd759b34dbc@uwinnipeg.ca> References: <22E7A439-EAF7-43E1-B7ED-AC0731F663EC@economics.gu.se> <244ecc218f6442699989dbd759b34dbc@uwinnipeg.ca> Message-ID: Phil. You are describing a time series application. However, most of the familiar analysis of IVs, and the standard "tests" apply to cross sections. Time series applications involving lagged effects and lagged variables are qualitatively different, since the distinction between endogenous and exogenous variables is more complex. Regards, Bill Greene On Tue, Aug 14, 2018 at 9:49 AM, Phil Cyrenne wrote: > Hello Professor Greene: > > I am sure you get swamped with econometric questions, but > > I have looked through your textbook and have this particular question about > > estimating a simultaneous equation system. > > With an endogenous regressor, in general an IV variable is needed, which > > as you know searching for an appropriate one can prove challenging. > > I have been under the impression that in the case of an endogenous > regressor, the > > common alternative has been to use the lagged value of the endogenous > regressor, > > since in general it is strongly correlated with the endogenous regresso > but is less likely to be correlated with the other endogenous regressor. I > have thought this was a relatively uncontroversial assumption. > > However, I suppose the lagged value of the endogenous regressor can be > viewed as an > > IV variable, but I have not come across papers where the standard IV tests > have been > > applied to the use of a lagged endogenous variable as a particular IV > variable. My question, are these tests really needed when a lagged > endogenous variable has been used as instrument? If no, why? If it > is, Is it a common practice? Finally is there > > a handy reference for this practice? > > Thanks in advance. > > > Phil Cyrenne > > ________________________________ > From: Limdep on behalf of William > Greene via Limdep > Sent: Tuesday, July 24, 2018 9:51:14 AM > To: Limdep and Nlogit Mailing List > Cc: William Greene > Subject: Re: [Limdep Nlogit List] WTp estimation > > exactly right. I agree > Bill Greene > > On Tue, Jul 24, 2018 at 10:01 AM, Fredrik Carlsson < > fredrik.carlsson at economics.gu.se> wrote: > > > You have one less parameter, so reduce the number of parameters in the > > labels part of the Wald-command > > > > Fredrik > > > > Skickat fr?n min iPhone > > > > > 24 juli 2018 kl. 15:53 skrev medard kakuru via Limdep < > > limdep at mailman.sydney.edu.au>: > > > > > > Thanks Bill. > > > I am facing another challenge. I use the wald command to estimate wtp > > after > > > estimating a mixed logit with all non-attribute coefficients given a > > normal > > > distribution and the syntax works. Since the standard deviation for one > > > attribute was not significant, I re-specify the model and make that > > > attribute non-random. Running the wald command after this specification > > > gives me an error message that "wrong number of start values - need one > > for > > > each label". What could be the problem? > > > > > > Thanks. > > > > > > Medard > > > > > > On Sun, Jul 22, 2018 at 12:35 AM, William Greene via Limdep < > > > limdep at mailman.sydney.edu.au> wrote: > > > > > >> Medard. WTP for all variants of the RP logit model is described in > > >> Chapter N29. That would be good for the S-MNL and GMXL. I'm > > >> not sure what the G-MNL model is. WTP is generally a ratio of > > >> coefficients, though in RP models, the WTP is computed for each > > >> individual. P values for LR statistics are generally not computed. > > >> I'm not sure what the P-value at 5% is. You can use CALC to look > > >> up critical values for chi-squared if you wish. The function > > >> CTB(prob,df) returns the critical value for probability prob with > > >> degrees of freedom df. > > >> Regards > > >> /Bill Greene > > >> > > >> On Fri, Jul 20, 2018 at 5:22 AM, medard kakuru via Limdep < > > >> limdep at mailman.sydney.edu.au> wrote: > > >> > > >>> Dear All, > > >>> Kindly help me with the syntax for wtp in preference space. I am > > running > > >>> S-MNL, G-MNL and the GMXL models and want to compare WTP estimates > for > > >> the > > >>> three models. I also want to compare these estimates with those > > obtained > > >>> from wtp-space model estimations. I understand that for a simple MXL > > >> model, > > >>> the syntax for wtp is just a ratio of attribute and cost > coefficients. > > Is > > >>> it the case with the models I am estimating? I expected to find the > > >> syntax > > >>> in the Nlogit manual (section N33) but it's not listed. > > >>> > > >>> > > >>> Secondly, I have seen the syntax for carrying out a loglikelihood > ratio > > >>> test and the syntax gives the LR statistic. How do i go further to > get > > a > > >>> p-value (say at 5%) for the statistic? > > >>> > > >>> Thanks in advance. > > >>> > > >>> Regards, Medard > > >>> _______________________________________________ > > >>> 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/Y7iMCVAGXPt9JOqGSGnjGp?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 > > >> > > >> > > > _______________________________________________ > > > Limdep site list > > > Limdep at mailman.sydney.edu.au > > > http://limdep.itls.usyd.edu.au > > > > > _______________________________________________ > > 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/Y7iMCVAGXPt9JOqGSGnjGp?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 > > _______________________________________________ > 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/Y7iMCVAGXPt9JOqGSGnjGp?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 p.cyrenne at uwinnipeg.ca Wed Aug 15 01:41:24 2018 From: p.cyrenne at uwinnipeg.ca (Phil Cyrenne) Date: Tue, 14 Aug 2018 15:41:24 +0000 Subject: [Limdep Nlogit List] WTp estimation In-Reply-To: References: <22E7A439-EAF7-43E1-B7ED-AC0731F663EC@economics.gu.se> <244ecc218f6442699989dbd759b34dbc@uwinnipeg.ca>, Message-ID: Thanks Professor Greene Over the years I have found your textbooks very helpful and informative All the best Phil Cyrenne ________________________________ From: Limdep on behalf of William Greene via Limdep Sent: Tuesday, August 14, 2018 10:15:45 AM To: Limdep and Nlogit Mailing List Cc: William Greene Subject: Re: [Limdep Nlogit List] WTp estimation Phil. You are describing a time series application. However, most of the familiar analysis of IVs, and the standard "tests" apply to cross sections. Time series applications involving lagged effects and lagged variables are qualitatively different, since the distinction between endogenous and exogenous variables is more complex. Regards, Bill Greene On Tue, Aug 14, 2018 at 9:49 AM, Phil Cyrenne wrote: > Hello Professor Greene: > > I am sure you get swamped with econometric questions, but > > I have looked through your textbook and have this particular question about > > estimating a simultaneous equation system. > > With an endogenous regressor, in general an IV variable is needed, which > > as you know searching for an appropriate one can prove challenging. > > I have been under the impression that in the case of an endogenous > regressor, the > > common alternative has been to use the lagged value of the endogenous > regressor, > > since in general it is strongly correlated with the endogenous regresso > but is less likely to be correlated with the other endogenous regressor. I > have thought this was a relatively uncontroversial assumption. > > However, I suppose the lagged value of the endogenous regressor can be > viewed as an > > IV variable, but I have not come across papers where the standard IV tests > have been > > applied to the use of a lagged endogenous variable as a particular IV > variable. My question, are these tests really needed when a lagged > endogenous variable has been used as instrument? If no, why? If it > is, Is it a common practice? Finally is there > > a handy reference for this practice? > > Thanks in advance. > > > Phil Cyrenne > > ________________________________ > From: Limdep on behalf of William > Greene via Limdep > Sent: Tuesday, July 24, 2018 9:51:14 AM > To: Limdep and Nlogit Mailing List > Cc: William Greene > Subject: Re: [Limdep Nlogit List] WTp estimation > > exactly right. I agree > Bill Greene > > On Tue, Jul 24, 2018 at 10:01 AM, Fredrik Carlsson < > fredrik.carlsson at economics.gu.se> wrote: > > > You have one less parameter, so reduce the number of parameters in the > > labels part of the Wald-command > > > > Fredrik > > > > Skickat fr?n min iPhone > > > > > 24 juli 2018 kl. 15:53 skrev medard kakuru via Limdep < > > limdep at mailman.sydney.edu.au>: > > > > > > Thanks Bill. > > > I am facing another challenge. I use the wald command to estimate wtp > > after > > > estimating a mixed logit with all non-attribute coefficients given a > > normal > > > distribution and the syntax works. Since the standard deviation for one > > > attribute was not significant, I re-specify the model and make that > > > attribute non-random. Running the wald command after this specification > > > gives me an error message that "wrong number of start values - need one > > for > > > each label". What could be the problem? > > > > > > Thanks. > > > > > > Medard > > > > > > On Sun, Jul 22, 2018 at 12:35 AM, William Greene via Limdep < > > > limdep at mailman.sydney.edu.au> wrote: > > > > > >> Medard. WTP for all variants of the RP logit model is described in > > >> Chapter N29. That would be good for the S-MNL and GMXL. I'm > > >> not sure what the G-MNL model is. WTP is generally a ratio of > > >> coefficients, though in RP models, the WTP is computed for each > > >> individual. P values for LR statistics are generally not computed. > > >> I'm not sure what the P-value at 5% is. You can use CALC to look > > >> up critical values for chi-squared if you wish. The function > > >> CTB(prob,df) returns the critical value for probability prob with > > >> degrees of freedom df. > > >> Regards > > >> /Bill Greene > > >> > > >> On Fri, Jul 20, 2018 at 5:22 AM, medard kakuru via Limdep < > > >> limdep at mailman.sydney.edu.au> wrote: > > >> > > >>> Dear All, > > >>> Kindly help me with the syntax for wtp in preference space. I am > > running > > >>> S-MNL, G-MNL and the GMXL models and want to compare WTP estimates > for > > >> the > > >>> three models. I also want to compare these estimates with those > > obtained > > >>> from wtp-space model estimations. I understand that for a simple MXL > > >> model, > > >>> the syntax for wtp is just a ratio of attribute and cost > coefficients. > > Is > > >>> it the case with the models I am estimating? I expected to find the > > >> syntax > > >>> in the Nlogit manual (section N33) but it's not listed. > > >>> > > >>> > > >>> Secondly, I have seen the syntax for carrying out a loglikelihood > ratio > > >>> test and the syntax gives the LR statistic. How do i go further to > get > > a > > >>> p-value (say at 5%) for the statistic? > > >>> > > >>> Thanks in advance. > > >>> > > >>> Regards, Medard > > >>> _______________________________________________ > > >>> 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/zM_2ClxwB5CprpjgHG84VH?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 > > >> > > >> > > > _______________________________________________ > > > Limdep site list > > > Limdep at mailman.sydney.edu.au > > > http://limdep.itls.usyd.edu.au > > > > > _______________________________________________ > > 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/zM_2ClxwB5CprpjgHG84VH?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 > > _______________________________________________ > 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/zM_2ClxwB5CprpjgHG84VH?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