From fengxiaz at gmail.com Wed Mar 1 04:01:27 2017 From: fengxiaz at gmail.com (fengxia zhu) Date: Tue, 28 Feb 2017 12:01:27 -0500 Subject: [Limdep Nlogit List] Seeking help on an error message related to Hausman Test in random effects panel model Message-ID: Hi, I have encountered the following error when running a random effects panel model. Error: 425: REGR;PANEL. Could not invert VC matrix for Hausman test. I looked over the LIMDEP manual and could not find a way to resolve it. Below is the command I used: REGRESS ;Lhs=Y ;Rhs=ONE, x1, x2 ;group=NID1 ; pds=FREQ ;Panel ;Random $ I know that the estimation of the model is ok but needed the test to determine whether my final model should be a fixed effects or random effects model. Any help or suggestions will be highly appreciated! Sincerely, FZ From p.paramita at qut.edu.au Wed Mar 8 15:54:53 2017 From: p.paramita at qut.edu.au (Puteri Paramita) Date: Wed, 8 Mar 2017 04:54:53 +0000 Subject: [Limdep Nlogit List] queries: how to explain heterogeneity in Ordered Probit model with random parameters Message-ID: Hi all, At the moment, I am analysing a comprehensive data set of socio-economic variables and journey characteristics variables of urban train riders from in Australia. I am employing NLOGIT/ Limdep software to estimate an ordered probit model with some random parameters, especially using OPROBIT command. This ordered probit model aims to estimate satisfaction level towards the train fare (in the form of ordinal variable) as a function of socio-demographic (in the form of categorical variables, i.e. income level, employment, etc). A few of the independent categorical variables are random parameters following either lognormal or normal distribution. Therefore, I added fcn line command in the model. The OPROBIT commands with random parameters works well, and I am able to estimate the parameter coefficients. Now, I would like to know more about those random parameters, e.g. what are the socio-demographic and journey experience characteristics that affect the heterogeneity among different respondents? In particular, one of the random parameters is one-way train fare, so I would like to test whether income level and employment of respondent affects their view on train fare? Is there any additional command line that I can insert to OPROBIT commands in order to explore the heterogeneity of its random parameters? Any other way that you could suggest to explore and explain the heterogeneity of the random parameters? Looking forward to hearing back from you soon. Thank you. Kind Regards, Puteri Paramita BSc.(Hons.) Statistics and Mathematics, MSc. Transport Planning Transportation Group | School of Civil Engineering and Built Environment Science and Engineering Faculty | Queensland University of Technology Room S831, Level 8, S Block, Gardens Point Campus | 2 George St, Brisbane Qld 4000 mail GPO Box 2434, Brisbane Qld 4001, Australia mobile +61 434 804 415 | email p.paramita at qut.edu.au | www.qut.edu.au From richard.turner at imarketresearch.com Thu Mar 9 02:48:30 2017 From: richard.turner at imarketresearch.com (Richard Turner) Date: Wed, 8 Mar 2017 10:48:30 -0500 Subject: [Limdep Nlogit List] Should mixed logit (ML) be able to replicate base choice shares and should MLs be used for forecasts? Message-ID: Greetings, I have a couple of questions. I would greatly appreciate some commentary on them. Thanks in advance for any answers. 1. Should mixed logit with alternative specific constants and estimated individual parameters be able to replicate the "base" choice shares (i.e. choice shares in the data) within the data? Alternative specific constants have this affect in multinomial logit choice models and I was wondering if the same applied to mixed logit. *In a mixed logit model that I estimated, I calculated shares using the fixed coefficients and the individual level parameters, but the shares were a bit off from the actual shares in the data*. I'm not sure if this would make a difference but, about 70% of my variables in the multinomial logit model to create my starting values were significant, but none of the parameters in my ML were. The ML parameters did not seem strange in magnitude though. 2. Should ML with estimated individual parameters be used for forecasts? I've read that they should not, but I am not sure why. Shouldn't the individual parameters, if accurate, create accurate forecasts for the population? If not, why should one trust that the distribution of parameters was estimated correctly in the first place. Regards, Richard From acsantanna at ksu.edu Thu Mar 9 03:02:07 2017 From: acsantanna at ksu.edu (Ana Claudia Sant'Anna) Date: Wed, 8 Mar 2017 16:02:07 +0000 Subject: [Limdep Nlogit List] Calculating Standard Errors for Arc Elasticities Message-ID: Dear list members, Following the "Applied Choice Analysis" book 2nd Edition, I am estimating the arc elasticities for dummy attributes in a multinomial logit model using the simulations code, scenario and arc (p. 512-526). My question is how to calculate the standard errors for these arc elasticities. Here is the code I am using and the results I am getting: |-> RPLOGIT ; lhs = choice, cset, alti ; Choices = lease, partner, supply, out, stop ; Model: U(lease) = a0 + a1*rate + a2*risk + a3*length + a11*more1/ U(partner) = b0 + b2*risk + b3*length + b4* sharepay + b5*trs + b11*more1/ U(supply) = c0 + c2*risk + c3*length + c5*trs + c6*plant + c7*harv + c8*delivery + c9*mbuyall + c11*more1/ U(stop) = d0 + d11*more1 ; Full;pwt ; RPM =well, relate ; FCN = a0(n), b0(n), c0(n), d0(n) ; Halton ; Pts = 1000 ; Pds = 6 ; Maxit = 1000 ; Alg = BFGS ; tlb =1e-20 ; tlg = 1e-6 ; tlf = 1e-20 ; output = 1 ;simulation;arc ;scenario: mbuyall(supply)=0 & mbuyall(supply)=1 ; Parameters ;export=tables$ +---------------------------------------------+ | Random Parameters Logit Model | | Model Simulation Using Previous Estimates | | Number of observations 648 | +---------------------------------------------+ +------------------------------------------------------+ |Simulations of Probability Model | |Model: Random Parameters Logit Model | |Simulated choice set may be a subset of the choices. | |Number of individuals is the probability times the | |number of observations in the simulated sample. | |Column totals may be affected by rounding error. | |The model used was simulated with 648 observations.| +------------------------------------------------------+ ---------------------------------------------------------------------------- Estimated Arc Elasticities Based on the Specified Scenario. Rows in the table report 0.00 if the indicated attribute did not change in the scenario or if the average probability or average attribute was zero in the sample. Estimated values are averaged over all individuals used in the simulation. Rows of the table in which no changes took place are not shown. ---------------------------------------------------------------------------- Attr Changed in | Change in Probability of Alternative ---------------------------------------------------------------------------- Choice SUPPLY | LEASE PARTNER SUPPLY OUT STOP x = C9 | -.012 -.014 .178 -.025 -.002 ---------------------------------------------------------------------------- ------------------------------------------------------------------------- Specification of scenario 1 is: Attribute Alternatives affected Change type Value --------- ------------------------------- ------------------- --------- MBUYALL SUPPLY Fix at new value .000 ------------------------------------------------------------------------- The simulator located 648 observations for this scenario. Simulated Probabilities (shares) for this scenario: +----------+--------------+--------------+------------------+ |Choice | Base | Scenario | Scenario - Base | | |%Share Number |%Share Number |ChgShare ChgNumber| +----------+--------------+--------------+------------------+ |LEASE | 18.968 123 | 19.443 126 | .475% 3 | |PARTNER | 21.290 138 | 21.913 142 | .623% 4 | |SUPPLY | 11.218 73 | 8.129 53 | -3.089% -20 | |OUT | 35.834 232 | 37.765 245 | 1.930% 13 | |STOP | 12.689 82 | 12.749 83 | .061% 1 | |Total |100.000 648 |100.000 649 | .000% 1 | +----------+--------------+--------------+------------------+ ------------------------------------------------------------------------- Specification of scenario 2 is: Attribute Alternatives affected Change type Value --------- ------------------------------- ------------------- --------- MBUYALL SUPPLY Fix at new value 1.000 ------------------------------------------------------------------------- The simulator located 648 observations for this scenario. Simulated Probabilities (shares) for this scenario: +----------+--------------+--------------+------------------+ |Choice | Base | Scenario | Scenario - Base | | |%Share Number |%Share Number |ChgShare ChgNumber| +----------+--------------+--------------+------------------+ |LEASE | 18.968 123 | 18.407 119 | -.562% -4 | |PARTNER | 21.290 138 | 20.636 134 | -.654% -4 | |SUPPLY | 11.218 73 | 14.602 95 | 3.384% 22 | |OUT | 35.834 232 | 33.730 219 | -2.105% -13 | |STOP | 12.689 82 | 12.626 82 | -.063% 0 | |Total |100.000 648 |100.000 649 | .000% 1 | +----------+--------------+--------------+------------------+ The simulator located 648 observations for this scenario. Pairwise Comparisons of Specified Scenarios Base for this comparison is scenario 1. Scenario for this comparison is scenario 2. +----------+--------------+--------------+------------------+ |Choice | Base | Scenario | Scenario - Base | | |%Share Number |%Share Number |ChgShare ChgNumber| +----------+--------------+--------------+------------------+ |LEASE | 19.443 126 | 18.407 119 | -1.037% -7 | |PARTNER | 21.913 142 | 20.636 134 | -1.277% -8 | |SUPPLY | 8.129 53 | 14.602 95 | 6.473% 42 | |OUT | 37.765 245 | 33.730 219 | -4.035% -26 | |STOP | 12.749 83 | 12.626 82 | -.124% -1 | |Total |100.000 649 |100.000 649 | .000% 0 | +----------+--------------+--------------+------------------+ Thank you for your help, Ana Claudia From wgreene at stern.nyu.edu Thu Mar 9 05:57:09 2017 From: wgreene at stern.nyu.edu (William Greene) Date: Wed, 8 Mar 2017 13:57:09 -0500 Subject: [Limdep Nlogit List] Should mixed logit (ML) be able to replicate base choice shares and should MLs be used for forecasts? In-Reply-To: References: Message-ID: Richard. 1. Mixed logit with ASCs will not exactly replicate the base choice shares. That result is a specific algebraic result for the MNL model. Other models, like the mixed logit will approximate the result, but only the MNL hits it exactly. 2. Not clear what you mean by "accurate forecasts for the population." But, the "individual" parameters are estimates of the conditional means of the parameter distribution. Averaged across individuals, you get an estimate of the unconditional means. If you are forecasting the sample totals, I would use the estimated population averages not the sum of the individual results. /Bill greene On Wed, Mar 8, 2017 at 10:48 AM, Richard Turner < richard.turner at imarketresearch.com> wrote: > Greetings, > > I have a couple of questions. I would greatly appreciate some commentary on > them. Thanks in advance for any answers. > > 1. Should mixed logit with alternative specific constants and estimated > individual parameters be able to replicate the "base" choice shares > (i.e. > choice shares in the data) within the data? Alternative specific > constants > have this affect in multinomial logit choice models and I was wondering > if > the same applied to mixed logit. *In a mixed logit model that I > estimated, I calculated shares using the fixed coefficients and the > individual level parameters, but the shares were a bit off from the > actual > shares in the data*. I'm not sure if this would make a difference but, > about 70% of my variables in the multinomial logit model to create my > starting values were significant, but none of the parameters in my ML > were. > The ML parameters did not seem strange in magnitude though. > 2. Should ML with estimated individual parameters be used for forecasts? > I've read that they should not, but I am not sure why. Shouldn't the > individual parameters, if accurate, create accurate forecasts for the > population? If not, why should one trust that the distribution of > parameters was estimated correctly in the first place. > > Regards, > > Richard > _______________________________________________ > Limdep site list > Limdep at limdep.itls.usyd.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: http://people.stern.nyu.edu/wgreene Email: wgreene at stern.nyu.edu Ph. +1.212.998.0876 From josaez at ugr.es Mon Mar 13 22:43:06 2017 From: josaez at ugr.es (josaez) Date: Mon, 13 Mar 2017 14:43:06 +0300 Subject: [Limdep Nlogit List] =?utf-8?q?any_ideas=3F?= Message-ID: <1701145686.20170313144306@ugr.es> Hey, I'm writing an article and I need some new ideas, can you take a look at it and tell me your ideas http://worth.franckhardy.com/5657 Warmest, josaez From rtay888 at gmail.com Wed Mar 15 16:04:57 2017 From: rtay888 at gmail.com (Richard Tay) Date: Wed, 15 Mar 2017 16:04:57 +1100 Subject: [Limdep Nlogit List] Error in random parameter logit model Message-ID: <001201d29d49$b38d48c0$1aa7da40$@gmail.com> I tried to run a random binary logit model and got an error saying that: Initial iterations cannot improve function.Status=3 Error: 805: Initial iterations cannot improve function.Status=3 Function= .11790081894D+05, at entry, .11790081894D+05 at exit What does this mean? In Nlogit, with the "logit; rpm $" command, the program will run the standard binary logit model first and use the estimate as start points for the rpm. The program run the standard logit without giving any errors but give an error when trying to estimate the rpm. Does this mean that there is no random coefficients and the standard binary logit model is OK? Cheers, Rich From wgreene at stern.nyu.edu Wed Mar 15 23:59:59 2017 From: wgreene at stern.nyu.edu (William Greene) Date: Wed, 15 Mar 2017 08:59:59 -0400 Subject: [Limdep Nlogit List] Error in random parameter logit model In-Reply-To: <001201d29d49$b38d48c0$1aa7da40$@gmail.com> References: <001201d29d49$b38d48c0$1aa7da40$@gmail.com> Message-ID: Rich. You have to tell the estimator what the random coefficients are. Your command is incomplete. It must include ;RPM ; Fcn = ... that gives the random coefficients. Please see the manual for explanation. /Bill Greene On Wed, Mar 15, 2017 at 1:04 AM, Richard Tay via Limdep < limdep at limdep.itls.usyd.edu.au> wrote: > I tried to run a random binary logit model and got an error saying that: > > > > Initial iterations cannot improve function.Status=3 > > Error: 805: Initial iterations cannot improve function.Status=3 > > Function= .11790081894D+05, at entry, .11790081894D+05 at exit > > > > What does this mean? In Nlogit, with the "logit; rpm $" command, the > program > will run the standard binary logit model first and use the estimate as > start > points for the rpm. The program run the standard logit without giving any > errors but give an error when trying to estimate the rpm. Does this mean > that there is no random coefficients and the standard binary logit model is > OK? > > > > Cheers, > > Rich > > > > > > _______________________________________________ > Limdep site list > Limdep at limdep.itls.usyd.edu.au > http://limdep.itls.usyd.edu.au > > -- William Greene Department of Economics Stern School of Business, New York University 44 West 4 St., 7-90 New York, NY, 10012 URL: http://people.stern.nyu.edu/wgreene Email: wgreene at stern.nyu.edu Ph. +1.212.998.0876 From gutamajale at gmail.com Thu Mar 16 16:57:49 2017 From: gutamajale at gmail.com (S Tarfasa) Date: Thu, 16 Mar 2017 01:57:49 -0400 Subject: [Limdep Nlogit List] Equality constrained latent class model Message-ID: Dear all, I need your help on how to estimate attribute non-attendance using equality constraint latent class model.I have seven variables(6 attributes and a monetary variable) i.e., X1,X2,X3,X4,X5,X6 and X7. I have 9 classes to use(PTs=103). class 1 none of the attributes igored X1 is ignored X2 is ignored X3 is ignored X4 is ignored X5 is ignore X6 is ignored X7 is ignored all is ignored. Constraining the coefficients of the ignored attributes/price,how do put the syntax using NLOGIT 5.? Thank you in advance for your help Solomon Tarfasa Assistant professor From djourdain at ait.asia Thu Mar 16 17:29:24 2017 From: djourdain at ait.asia (Damien Jourdain) Date: Thu, 16 Mar 2017 13:29:24 +0700 Subject: [Limdep Nlogit List] Equality constrained latent class model In-Reply-To: References: Message-ID: <000001d29e1e$aa6db650$ff4922f0$@cirad.fr> Dear Solomon, As far as I understand, you can either use the 2K formulation (well explained in the manual), or use a more flexible approach of imposing the constraints you deem necessary on the coefficients. The second solution is more flexible as you can impose different constraints on which of the coefficients should be zero in which class. You can even decide some classes ignore 2 or more attributes at the same time. I understand the two approaches are equivalent, but this needs to be confirmed by the developers? Here is an example with my own variables: LCLOGIT ; Choices = 1,2,3 ; Lhs = CHOICE ; Rhs = pay, mosq, waste, trail, quiet, fecos, asc ; Pds = csi ; Pts =9 ; Rst = pay, mosq, waste, trail, quiet, fecos, asc, 0, mosq, waste, trail, quiet, fecos, asc, pay,0, waste, trail, quiet, fecos, asc, pay, mosq, 0, trail, quiet, fecos, asc, pay, mosq, waste, 0, quiet, fecos, asc, pay, mosq, waste, trail, 0, fecos, asc, pay, mosq, waste, trail, quiet, 0, asc, pay, mosq, waste, trail, quiet, fecos, 0, 0, 0, 0, 0, 0, 0, 0 $ I hope this helps. Damien Jourdain Asian Institute of Technology. -----Message d'origine----- De?: Limdep [mailto:limdep-bounces at limdep.itls.usyd.edu.au] De la part de S Tarfasa via Limdep Envoy??: Thursday, March 16, 2017 12:58 PM ??: limdep at limdep.itls.usyd.edu.au Cc?: S Tarfasa Objet?: [Limdep Nlogit List] Equality constrained latent class model Dear all, I need your help on how to estimate attribute non-attendance using equality constraint latent class model.I have seven variables(6 attributes and a monetary variable) i.e., X1,X2,X3,X4,X5,X6 and X7. I have 9 classes to use(PTs=103). class 1 none of the attributes igored X1 is ignored X2 is ignored X3 is ignored X4 is ignored X5 is ignore X6 is ignored X7 is ignored all is ignored. Constraining the coefficients of the ignored attributes/price,how do put the syntax using NLOGIT 5.? Thank you in advance for your help Solomon Tarfasa Assistant professor _______________________________________________ Limdep site list Limdep at limdep.itls.usyd.edu.au http://limdep.itls.usyd.edu.au From gutamajale at gmail.com Fri Mar 17 08:32:33 2017 From: gutamajale at gmail.com (S Tarfasa) Date: Thu, 16 Mar 2017 17:32:33 -0400 Subject: [Limdep Nlogit List] Help to estimate equality constrained latent class model Message-ID: Dear All, Would anyone help me estimate equality constraint latent class model as I was challenged doing it? I have 9 classes where some respondents ignored one attribute,others all attributes and the remaining ignored none.I keep getting error when I run the following syntxa butI couldn't figure it out. Any help is highly appreciated. Error 451: Expected 27 specifications in RST/CML list. Found 72. Error 1071: Found an error in your ;RST setup. Please check. Error 1071: Found an error in your ;RST setup. Please check. Best, LCLOGIT ;LHS=CHOICE ;CHOICES=1,2,3 ;MODEL: U(1,2,3)=BPAY*PAYMENT+BCOV*COVER2+BCEM*CEME+BLAB*LABOR+BCAP*CAP1000+BDUR*DURATION+BRGVT*RGVT+ASC ;PTS=9 ;PDS=10 ;rst=bpay,bcov,bcem,blab,bcap,bdur,brgvt,ASC, bpay,bcov,bcem,blab,bcap,bdur,0,ASC, bpay,bcov,bcem,blab,bcap,0,brgvt,ASC, bpay,bcov,bcem,blab,0,bdur,brgvt,ASC, bpay,bcov,bcem,0,bcap,bdur,brgvt,ASC, bpay,bcov,0,blab,bcap,bdur,brgvt,ASC, bpay,0,bcem,blab,bcap,bdur,brgvt,ASC, 0,bcov,bcem,blab,bcap,bdur,brgvt,ASC, 0,0,0,0,0,0,0,0$ From gutamajale at gmail.com Fri Mar 17 14:19:37 2017 From: gutamajale at gmail.com (S Tarfasa) Date: Thu, 16 Mar 2017 23:19:37 -0400 Subject: [Limdep Nlogit List] Help to estimate equality constrained latent class model In-Reply-To: <00df01d29ecc$e5397270$afac5750$@tceagle.com> References: <00df01d29ecc$e5397270$afac5750$@tceagle.com> Message-ID: Dear sir, Thank you for your help. I already got the solution. On 16 Mar 2017 11:16 pm, "Thomas Eagle" wrote: > Why do you want to constrain all parameters across 8 of the classes, set a > different to zero in the first 8, and all parameters to be zero in the last > class? What kind of latent class model is that? The purpose of a latent > class choice model is to find different groups of respondents who have > differing parameter values across classes? Usually one fits an > unconstrained LC choice model and then imposes constraints to make the > model > more parsimonious. > > I can see some simplified constraints to force a type of mover-stayer > model, > but I cannot fathom the model you are trying to fit... > > Tom Eagle > > -----Original Message----- > From: Limdep [mailto:limdep-bounces at limdep.itls.usyd.edu.au] On Behalf Of > S > Tarfasa via Limdep > Sent: Thursday, March 16, 2017 2:33 PM > To: limdep at limdep.itls.usyd.edu.au > Cc: S Tarfasa > Subject: [Limdep Nlogit List] Help to estimate equality constrained latent > class model > > Dear All, > > Would anyone help me estimate equality constraint latent class model as I > was challenged doing it? > > I have 9 classes where some respondents ignored one attribute,others all > attributes and the remaining ignored none.I keep getting error when I run > the following syntxa butI couldn't figure it out. > Any help is highly appreciated. > > Error 451: Expected 27 specifications in RST/CML list. Found 72. > Error 1071: Found an error in your ;RST setup. Please check. > Error 1071: Found an error in your ;RST setup. Please check. > Best, > > > > > LCLOGIT > ;LHS=CHOICE > ;CHOICES=1,2,3 > ;MODEL: > U(1,2,3)=BPAY*PAYMENT+BCOV*COVER2+BCEM*CEME+BLAB*LABOR+ > BCAP*CAP1000+BDUR*DUR > ATION+BRGVT*RGVT+ASC > ;PTS=9 > ;PDS=10 > ;rst=bpay,bcov,bcem,blab,bcap,bdur,brgvt,ASC, > bpay,bcov,bcem,blab,bcap,bdur,0,ASC, > bpay,bcov,bcem,blab,bcap,0,brgvt,ASC, > bpay,bcov,bcem,blab,0,bdur,brgvt,ASC, > bpay,bcov,bcem,0,bcap,bdur,brgvt,ASC, > bpay,bcov,0,blab,bcap,bdur,brgvt,ASC, > bpay,0,bcem,blab,bcap,bdur,brgvt,ASC, > 0,bcov,bcem,blab,bcap,bdur,brgvt,ASC, > 0,0,0,0,0,0,0,0$ > _______________________________________________ > Limdep site list > Limdep at limdep.itls.usyd.edu.au > http://limdep.itls.usyd.edu.au > > From david.hensher at sydney.edu.au Fri Mar 17 14:22:20 2017 From: david.hensher at sydney.edu.au (David Hensher) Date: Fri, 17 Mar 2017 03:22:20 +0000 Subject: [Limdep Nlogit List] Help to estimate equality constrained latent class model In-Reply-To: References: <00df01d29ecc$e5397270$afac5750$@tceagle.com> Message-ID: <58CB566E.5060604@sydney.edu.au> These details are set out in the second edition of Hensher, Rose and Greene (2015) Applied Choice analysis, Cambridge Uni Press, pages 729-30. David On 17/03/2017 2:19 PM, S Tarfasa via Limdep wrote: > Dear sir, > > Thank you for your help. > I already got the solution. > > > On 16 Mar 2017 11:16 pm, "Thomas Eagle" wrote: > > >> Why do you want to constrain all parameters across 8 of the classes, set a >> different to zero in the first 8, and all parameters to be zero in the last >> class? What kind of latent class model is that? The purpose of a latent >> class choice model is to find different groups of respondents who have >> differing parameter values across classes? Usually one fits an >> unconstrained LC choice model and then imposes constraints to make the >> model >> more parsimonious. >> >> I can see some simplified constraints to force a type of mover-stayer >> model, >> but I cannot fathom the model you are trying to fit... >> >> Tom Eagle >> >> -----Original Message----- >> From: Limdep [mailto:limdep-bounces at limdep.itls.usyd.edu.au] On Behalf Of >> S >> Tarfasa via Limdep >> Sent: Thursday, March 16, 2017 2:33 PM >> To: limdep at limdep.itls.usyd.edu.au >> Cc: S Tarfasa >> Subject: [Limdep Nlogit List] Help to estimate equality constrained latent >> class model >> >> Dear All, >> >> Would anyone help me estimate equality constraint latent class model as I >> was challenged doing it? >> >> I have 9 classes where some respondents ignored one attribute,others all >> attributes and the remaining ignored none.I keep getting error when I run >> the following syntxa butI couldn't figure it out. >> Any help is highly appreciated. >> >> Error 451: Expected 27 specifications in RST/CML list. Found 72. >> Error 1071: Found an error in your ;RST setup. Please check. >> Error 1071: Found an error in your ;RST setup. Please check. >> Best, >> >> >> >> >> LCLOGIT >> ;LHS=CHOICE >> ;CHOICES=1,2,3 >> ;MODEL: >> U(1,2,3)=BPAY*PAYMENT+BCOV*COVER2+BCEM*CEME+BLAB*LABOR+ >> BCAP*CAP1000+BDUR*DUR >> ATION+BRGVT*RGVT+ASC >> ;PTS=9 >> ;PDS=10 >> ;rst=bpay,bcov,bcem,blab,bcap,bdur,brgvt,ASC, >> bpay,bcov,bcem,blab,bcap,bdur,0,ASC, >> bpay,bcov,bcem,blab,bcap,0,brgvt,ASC, >> bpay,bcov,bcem,blab,0,bdur,brgvt,ASC, >> bpay,bcov,bcem,0,bcap,bdur,brgvt,ASC, >> bpay,bcov,0,blab,bcap,bdur,brgvt,ASC, >> bpay,0,bcem,blab,bcap,bdur,brgvt,ASC, >> 0,bcov,bcem,blab,bcap,bdur,brgvt,ASC, >> 0,0,0,0,0,0,0,0$ >> _______________________________________________ >> Limdep site list >> Limdep at limdep.itls.usyd.edu.au >> http://limdep.itls.usyd.edu.au >> >> >> > _______________________________________________ > Limdep site list > Limdep at limdep.itls.usyd.edu.au > http://limdep.itls.usyd.edu.au > > > -- DAVID HENSHER FASSA, PhD| Professor and Founding Director Institute of Transport and Logistics Studies | The University of Sydney Business School THE UNIVERSITY OF SYDNEY Rm 201, Building H73| The University of Sydney | NSW | 2006 Street Address: 378 Abercrombie St, Darlington NSW 2008 T +61 2 9114 1871 | F +61 2 9114 1863 | M +61 418 433 057 E David.Hensher at sydney.edu.au | W sydney.edu.au/business/itls Celebrating 25 years of ITLS: 1991-2016 http://youtu.be/s2D0T1crZwY ERA Rank 5 (Transportation and Freight Services) Co-Founder of the International Conference Series on Competition and Ownership of Land Passenger Transport (The 'Thredbo' Series) http://www.thredbo-conference-series.org/ Join the ITLS group on LinkedIn Second edition of Applied Choice Analysis now available at www.cambridge.org/9781107465923 CRICOS 00026A This email plus any attachments to it are confidential. Any unauthorised use is strictly prohibited. If you receive this email in error, please delete it and any attachments. Please think of our environment and only print this e-mail if necessary. From teagle at tceagle.com Fri Mar 17 14:16:36 2017 From: teagle at tceagle.com (Thomas Eagle) Date: Thu, 16 Mar 2017 20:16:36 -0700 Subject: [Limdep Nlogit List] Help to estimate equality constrained latent class model In-Reply-To: References: Message-ID: <00df01d29ecc$e5397270$afac5750$@tceagle.com> Why do you want to constrain all parameters across 8 of the classes, set a different to zero in the first 8, and all parameters to be zero in the last class? What kind of latent class model is that? The purpose of a latent class choice model is to find different groups of respondents who have differing parameter values across classes? Usually one fits an unconstrained LC choice model and then imposes constraints to make the model more parsimonious. I can see some simplified constraints to force a type of mover-stayer model, but I cannot fathom the model you are trying to fit... Tom Eagle -----Original Message----- From: Limdep [mailto:limdep-bounces at limdep.itls.usyd.edu.au] On Behalf Of S Tarfasa via Limdep Sent: Thursday, March 16, 2017 2:33 PM To: limdep at limdep.itls.usyd.edu.au Cc: S Tarfasa Subject: [Limdep Nlogit List] Help to estimate equality constrained latent class model Dear All, Would anyone help me estimate equality constraint latent class model as I was challenged doing it? I have 9 classes where some respondents ignored one attribute,others all attributes and the remaining ignored none.I keep getting error when I run the following syntxa butI couldn't figure it out. Any help is highly appreciated. Error 451: Expected 27 specifications in RST/CML list. Found 72. Error 1071: Found an error in your ;RST setup. Please check. Error 1071: Found an error in your ;RST setup. Please check. Best, LCLOGIT ;LHS=CHOICE ;CHOICES=1,2,3 ;MODEL: U(1,2,3)=BPAY*PAYMENT+BCOV*COVER2+BCEM*CEME+BLAB*LABOR+BCAP*CAP1000+BDUR*DUR ATION+BRGVT*RGVT+ASC ;PTS=9 ;PDS=10 ;rst=bpay,bcov,bcem,blab,bcap,bdur,brgvt,ASC, bpay,bcov,bcem,blab,bcap,bdur,0,ASC, bpay,bcov,bcem,blab,bcap,0,brgvt,ASC, bpay,bcov,bcem,blab,0,bdur,brgvt,ASC, bpay,bcov,bcem,0,bcap,bdur,brgvt,ASC, bpay,bcov,0,blab,bcap,bdur,brgvt,ASC, bpay,0,bcem,blab,bcap,bdur,brgvt,ASC, 0,bcov,bcem,blab,bcap,bdur,brgvt,ASC, 0,0,0,0,0,0,0,0$ _______________________________________________ Limdep site list Limdep at limdep.itls.usyd.edu.au http://limdep.itls.usyd.edu.au From SXB1022 at student.bham.ac.uk Wed Mar 22 21:18:02 2017 From: SXB1022 at student.bham.ac.uk (Saul Basurto Hernandez) Date: Wed, 22 Mar 2017 10:18:02 +0000 Subject: [Limdep Nlogit List] Nested Logit sequential estimator Message-ID: <09629F44AB19A249A1AFEDC0EE0614DCF137DF05@EX12.adf.bham.ac.uk> Dear all, I am currently working on farmers' crop and animals selections using the Nestel Logit model. As it comprises 31 alternatives at the bottom level and I have more than 200,000 cases, I want to estimate this model using the Sequential Estimator first. I am thinking about buying LIMDEP and NLOGIT packages in order to estimate it but, I would like to know whether NLOGIT is able to correct standard errors in the upper levels of the tree or not (without typing my own code). Thanks in advance, Saul From cem.payasli at emu.edu.tr Thu Mar 23 20:45:43 2017 From: cem.payasli at emu.edu.tr (Cem Payasli) Date: Thu, 23 Mar 2017 12:45:43 +0300 Subject: [Limdep Nlogit List] Industry Classifications in Random Parameter Model Message-ID: <000001d2a3ba$41412a80$c3c37f80$@emu.edu.tr> Greetings, I want to estimate a Random Parameter (NegBin-P) panel count model with industrial heterogenity effects accounted for by various type of economic activity classifications such 1) NACE2, 2) ISIC4 , 3) ISB3D and finally standard Industry classification in the original dataset. Specifically my Limdep command would look like : NEGBIN;Lhs=PAT;Rhs=LRD,LRD1,LRD2,LNS,LCAP,RDORD;PANEL;Robust;RPM=NACE2;Fcn = LCAP(n),LNS(n);Correlated;halton;Pts=25;Model=NBP$ In each round NACE2 will be replaced by ISIC4 and then by ISB3D and finally Industry categories. These classifications have 212, 32, 39 and 40 records. I have seen Random parameters example based on Bertscheck and Lechner (1998) study which defines an innovation binary logit variable as below: ? Random parameters with industry heterogeneity (From Greene's Lab session slide) ? Examine effect of industry heterogeneity. Sample ; All $ Logit ; Lhs = IP ; Rhs = One,IMUM,FDIUM,SP,LogSales ; Pds = 5 ; RPM = InvGood, RawMtl ; Halton ; Pts = 15 ; Cor ; Fcn = One(n),IMUM(n),FDIUM(n) ; Marginal ; Parameters $ Create; Bimum = beta_i(firm,2) $ Regress ; Lhs = Bimum ; Rhs = one,InvGood,RawMtl $ Because InvGood and RawMtl (Investment goods and Raw materials) are defined as dummy variables, their range in RPM specification is restricted to 0 and 1 only. Since my own RPM predictors are not dummies but factor variables (in Stata jargon) the only way to incorporate them seems to create dummies via CREATE;Expand(NACE2)=N1,N2,N3......so on.. However, that would be formidable number of dummies for some classifications. Is there any other way around ? Thanks for any suggestions. Cem Payasl?o?lu Department of Economics Eastern Mediterranaen University Magusa, TRNC, Mersin 10 Turkey