From wgreene at stern.nyu.edu  Tue Aug  1 03:03:06 2017
From: wgreene at stern.nyu.edu (William Greene)
Date: Mon, 31 Jul 2017 13:03:06 -0400
Subject: [Limdep Nlogit List] Help required on silly matter
In-Reply-To: <SY3PR01MB1097823E4BB9B1779321260BEBB20@SY3PR01MB1097.ausprd01.prod.outlook.com>
References: <SY3PR01MB1097823E4BB9B1779321260BEBB20@SY3PR01MB1097.ausprd01.prod.outlook.com>
Message-ID: <CAEJPMR5mTj-SNxvEWR39emhhKcz_x7JcFrZM5AJEoEBnPF0-vw@mail.gmail.com>

Yasir. The problem is the command. But, since you did not show the command,
we cannot
diagnose this.
/Bill Greene

On Mon, Jul 31, 2017 at 5:18 AM, Yasir Ali via Limdep <
limdep at mailman.sydney.edu.au> wrote:

> Hello,
>
> I am quite new to NLOGIT and trying to run MNL model.  While running the
> model, I am getting this error "Error 1003: A choice label appears more
> than once in the tree spec." I am unable to figure out that where is the
> issue? Either in the code for running MNL or in the data?
> I have checked the data and in the time series data, only one choice is
> selected at one instance. Can anyone please help me to resolve this issue?
> Thank you.
>
> Cheers,
> Yasir
>
> _______________________________________________
> 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: http://people.stern.nyu.edu/wgreene
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 M.A.Young at soton.ac.uk  Tue Aug  1 03:37:22 2017
From: M.A.Young at soton.ac.uk (Young M.A.)
Date: Mon, 31 Jul 2017 17:37:22 +0000
Subject: [Limdep Nlogit List] Error components ML
Message-ID: <45211921C1E44145B3C8B1291124700776629620@SRV00360.soton.ac.uk>


Hi,

As far as I can tell from the NLOGIT manual, the only specification available for the error components model is the one that uses the tree structure (nested) analogy.

I would like to specify the correlation based on a variable, which would seem to be  what the error components interpretation of ML should allow.  In my case I want this to be the distance between each pair of alternatives (or some function of this distance).

Is this possible using NLOGIT?

Thanks

Marcus

From y2.ali at qut.edu.au  Tue Aug  1 07:47:33 2017
From: y2.ali at qut.edu.au (Yasir Ali)
Date: Mon, 31 Jul 2017 21:47:33 +0000
Subject: [Limdep Nlogit List] Help required on silly matter
In-Reply-To: <CAEJPMR5mTj-SNxvEWR39emhhKcz_x7JcFrZM5AJEoEBnPF0-vw@mail.gmail.com>
References: <SY3PR01MB1097823E4BB9B1779321260BEBB20@SY3PR01MB1097.ausprd01.prod.outlook.com>
	<CAEJPMR5mTj-SNxvEWR39emhhKcz_x7JcFrZM5AJEoEBnPF0-vw@mail.gmail.com>
Message-ID: <SY3PR01MB109728D8A8860D6D2FD4B1F7EBB20@SY3PR01MB1097.ausprd01.prod.outlook.com>

Sorry Prof. Bill for not uploading the command. Please see the command below.

{
Nlogit
;lhs = choice,cset,altij
;choices = St-1,St-2,St-3,St-4,St-4,St-5,St-6
;checkdata
;model:
U(St-1) = A0 + A1*Acc_A_M + A2*Acc_FV + A3*Acc_lead + A4*Acc_M_A /
U(St-2) = A5 + A5*Acc_A_M + A7*Acc_FV + A8*Acc_lead + A9*Acc_M_A /
U(St-3) = A10 + A11*Acc_A_M + A12*Acc_FV + A13*Acc_lead + A14*Acc_M_A /
U(St-4) = A15 + A16*Acc_A_M + A17*Acc_M_A /
U(St-5) = A18 + A19*Acc_A_M + A20*Acc_FV + A21*Acc_lead + A22*Acc_M_A /
U(St-6) = 	     A23*Acc_A_M + A24*Acc_M_A 
$
}

Yasir


-----Original Message-----
From: Limdep [mailto:limdep-bounces at mailman.sydney.edu.au] On Behalf Of William Greene
Sent: Tuesday, 1 August 2017 3:03 AM
To: Limdep and Nlogit Mailing List <limdep at mailman.sydney.edu.au>
Subject: Re: [Limdep Nlogit List] Help required on silly matter

Yasir. The problem is the command. But, since you did not show the command, we cannot diagnose this.
/Bill Greene

On Mon, Jul 31, 2017 at 5:18 AM, Yasir Ali via Limdep < limdep at mailman.sydney.edu.au> wrote:

> Hello,
>
> I am quite new to NLOGIT and trying to run MNL model.  While running 
> the model, I am getting this error "Error 1003: A choice label appears 
> more than once in the tree spec." I am unable to figure out that where 
> is the issue? Either in the code for running MNL or in the data?
> I have checked the data and in the time series data, only one choice 
> is selected at one instance. Can anyone please help me to resolve this issue?
> Thank you.
>
> Cheers,
> Yasir
>
> _______________________________________________
> 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: http://people.stern.nyu.edu/wgreene
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 y2.ali at qut.edu.au  Tue Aug  1 08:47:39 2017
From: y2.ali at qut.edu.au (Yasir Ali)
Date: Mon, 31 Jul 2017 22:47:39 +0000
Subject: [Limdep Nlogit List] Help required on silly matter
In-Reply-To: <CAEJPMR5mTj-SNxvEWR39emhhKcz_x7JcFrZM5AJEoEBnPF0-vw@mail.gmail.com>
References: <SY3PR01MB1097823E4BB9B1779321260BEBB20@SY3PR01MB1097.ausprd01.prod.outlook.com>
	<CAEJPMR5mTj-SNxvEWR39emhhKcz_x7JcFrZM5AJEoEBnPF0-vw@mail.gmail.com>
Message-ID: <SY3PR01MB10972E47609918CE8448281AEBB20@SY3PR01MB1097.ausprd01.prod.outlook.com>

I just found issue in my code. Thanks Prof. Bill for prompt resonse.

-----Original Message-----
From: Limdep [mailto:limdep-bounces at mailman.sydney.edu.au] On Behalf Of William Greene
Sent: Tuesday, 1 August 2017 3:03 AM
To: Limdep and Nlogit Mailing List <limdep at mailman.sydney.edu.au>
Subject: Re: [Limdep Nlogit List] Help required on silly matter

Yasir. The problem is the command. But, since you did not show the command, we cannot diagnose this.
/Bill Greene

On Mon, Jul 31, 2017 at 5:18 AM, Yasir Ali via Limdep < limdep at mailman.sydney.edu.au> wrote:

> Hello,
>
> I am quite new to NLOGIT and trying to run MNL model.  While running 
> the model, I am getting this error "Error 1003: A choice label appears 
> more than once in the tree spec." I am unable to figure out that where 
> is the issue? Either in the code for running MNL or in the data?
> I have checked the data and in the time series data, only one choice 
> is selected at one instance. Can anyone please help me to resolve this issue?
> Thank you.
>
> Cheers,
> Yasir
>
> _______________________________________________
> 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: http://people.stern.nyu.edu/wgreene
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 M.A.Young at soton.ac.uk  Thu Aug 10 02:21:06 2017
From: M.A.Young at soton.ac.uk (Young M.A.)
Date: Wed, 9 Aug 2017 16:21:06 +0000
Subject: [Limdep Nlogit List] Error components ML
In-Reply-To: <45211921C1E44145B3C8B1291124700776629620@SRV00360.soton.ac.uk>
References: <45211921C1E44145B3C8B1291124700776629620@SRV00360.soton.ac.uk>
Message-ID: <45211921C1E44145B3C8B129112470077662E0C1@SRV00049.soton.ac.uk>

Hi,

Can no one provide an answer or any guidance on this question?

Marcus

-----Original Message-----
From: Limdep [mailto:limdep-bounces at mailman.sydney.edu.au] On Behalf Of Young M.A.
Sent: 31 July 2017 18:37
To: Limdep at limdep.itls.usyd.edu.au
Subject: [Limdep Nlogit List] Error components ML


Hi,

As far as I can tell from the NLOGIT manual, the only specification available for the error components model is the one that uses the tree structure (nested) analogy.

I would like to specify the correlation based on a variable, which would seem to be  what the error components interpretation of ML should allow.  In my case I want this to be the distance between each pair of alternatives (or some function of this distance).

Is this possible using NLOGIT?

Thanks

Marcus
_______________________________________________
Limdep site list
Limdep at mailman.sydney.edu.au
http://limdep.itls.usyd.edu.au


From y2.ali at qut.edu.au  Thu Aug 10 10:32:37 2017
From: y2.ali at qut.edu.au (Yasir Ali)
Date: Thu, 10 Aug 2017 00:32:37 +0000
Subject: [Limdep Nlogit List] PDS for time-series data
Message-ID: <SY3PR01MB1097E3DC3EECCAEF40C894C7EB880@SY3PR01MB1097.ausprd01.prod.outlook.com>

Hello everyone,

I am using time-series data in which I have a response of participant at series of values. The time is not fixed, so I have a different set of values for one participant and similarly different for another participant. Is there any way to capture this repetition of values in MNL model?
I know about PDS in which we define a number of repetition, in my case, repetition varies.
Any help would be appreciated. Thanks.

Regards,
Yasir

From dlortega at anr.msu.edu  Wed Aug 16 00:29:13 2017
From: dlortega at anr.msu.edu (Ortega, David)
Date: Tue, 15 Aug 2017 14:29:13 +0000
Subject: [Limdep Nlogit List] Convergence problem in WTP-space estimation
Message-ID: <FFFBD1B9-6D93-4CE1-A3D2-27928DE5F944@anr.msu.edu>

Hello,

I am having trouble getting an RPL model to converge in WTP-space (the model reaches 500 iterations without converging). However, if I run the same RPL model in preference space and then run that model in WTP-space, the model converges. Can anyone provide insight as to what is causing this? This is happening in more than one dataset. I am using Nlogit5.0.

Thanks, David



From RJohnston at clarku.edu  Wed Aug 16 01:34:29 2017
From: RJohnston at clarku.edu (Robert Johnston)
Date: Tue, 15 Aug 2017 15:34:29 +0000
Subject: [Limdep Nlogit List] Convergence problem in WTP-space estimation
In-Reply-To: <FFFBD1B9-6D93-4CE1-A3D2-27928DE5F944@anr.msu.edu>
References: <FFFBD1B9-6D93-4CE1-A3D2-27928DE5F944@anr.msu.edu>
Message-ID: <MWHPR03MB28145812BE33CA58C63D32EAD58D0@MWHPR03MB2814.namprd03.prod.outlook.com>

David,

This is a common occurrence and is expected given differences between the two models.  When estimated in WTP-space, the utility function is nonlinear-in-the-parameters.  This leads to computational and convergence challenges beyond those encountered in an otherwise identical (linear-in-the-parameters) preference-space utility function.  It relatively common for WTP-space estimation to encounter local optima, flat spots in the likelihood surface, etc.  You might consider diagnosing your model using different starting values or convergence criteria, or try alternative optimization algorithms.  These can all be adjusted in NLOGIT.  Even under the best of conditions, WTP-space models sometimes require many iterations to converge.

Robert J. Johnston
Director, George Perkins Marsh Institute
Professor, Department of Economics
Clark University
950 Main St.
Worcester, MA 01610
Phone: (508) 751-4619
Fax: (508) 751-4600
Email: rjohnston at clarku.edu
https://protect-au.mimecast.com/s/z4nRBmUpn2pID?domain=clarku.edu


-----Original Message-----
From: Limdep [mailto:limdep-bounces at mailman.sydney.edu.au] On Behalf Of Ortega, David
Sent: Tuesday, August 15, 2017 10:29 AM
To: limdep at mailman.sydney.edu.au
Subject: [Limdep Nlogit List] Convergence problem in WTP-space estimation

Hello,

I am having trouble getting an RPL model to converge in WTP-space (the model reaches 500 iterations without converging). However, if I run the same RPL model in preference space and then run that model in WTP-space, the model converges. Can anyone provide insight as to what is causing this? This is happening in more than one dataset. I am using Nlogit5.0.

Thanks, David


_______________________________________________
Limdep site list
Limdep at mailman.sydney.edu.au
http://limdep.itls.usyd.edu.au


From zhoufan88 at hotmail.com  Thu Aug 17 18:49:39 2017
From: zhoufan88 at hotmail.com (FAN Zhou)
Date: Thu, 17 Aug 2017 08:49:39 +0000
Subject: [Limdep Nlogit List] Interpretation of Mixed logit model
Message-ID: <KL1PR04MB1221A2804597FB393C036709AE830@KL1PR04MB1221.apcprd04.prod.outlook.com>

Hi all,


I have run a Mixed logit model with Nlogit 6. But I am confused by the results.


I was wondering how to interpret the insignificant mean but significant standard distribution for a random parameter. For example, if I set "cost" as random parameter, the mean 'B_COST' is insignificant but its associated standard deviation "NsCOST" is significant. Does results like this make sense?


I appreciate any help.


Cheers,

Fan

From miq at wne.uw.edu.pl  Thu Aug 17 19:01:28 2017
From: miq at wne.uw.edu.pl (=?UTF-8?Q?Miko=c5=82aj_Czajkowski?=)
Date: Thu, 17 Aug 2017 11:01:28 +0200
Subject: [Limdep Nlogit List] Interpretation of Mixed logit model
In-Reply-To: <KL1PR04MB1221A2804597FB393C036709AE830@KL1PR04MB1221.apcprd04.prod.outlook.com>
References: <KL1PR04MB1221A2804597FB393C036709AE830@KL1PR04MB1221.apcprd04.prod.outlook.com>
Message-ID: <38232f4b-1321-3595-46e1-1b6fa7ff3035@wne.uw.edu.pl>


Dear Fan,

in mixed logit a parameter is described using a parametric probability 
distribution. This distribution could have a mean that is close to zero 
(not significantly different) and some non-zero standard deviation 
(significantly different from 0). This is a fine result meaning that on 
average (mean) the effect is close to 0, but there is some heterogeneity 
around this. On the other hand, if standard deviation is not 
significantly different from 0 - you might as well model such a 
parameter as fixed, as heterogeneity around the mean is not significant.

I cannot say if it makes sense for your variables / model, but 
technically such a result is fine.

Best,
Mik



On 2017-08-17 10:49, FAN Zhou via Limdep wrote:
> Hi all,
>
>
> I have run a Mixed logit model with Nlogit 6. But I am confused by the results.
>
>
> I was wondering how to interpret the insignificant mean but significant standard distribution for a random parameter. For example, if I set "cost" as random parameter, the mean 'B_COST' is insignificant but its associated standard deviation "NsCOST" is significant. Does results like this make sense?
>
>
> I appreciate any help.
>
>
> Cheers,
>
> Fan
> _______________________________________________
> Limdep site list
> Limdep at mailman.sydney.edu.au
> http://limdep.itls.usyd.edu.au
>


From medakseth at gmail.com  Thu Aug 17 19:10:29 2017
From: medakseth at gmail.com (medard kakuru)
Date: Thu, 17 Aug 2017 12:10:29 +0300
Subject: [Limdep Nlogit List] Interpretation of Mixed logit model
In-Reply-To: <KL1PR04MB1221A2804597FB393C036709AE830@KL1PR04MB1221.apcprd04.prod.outlook.com>
References: <KL1PR04MB1221A2804597FB393C036709AE830@KL1PR04MB1221.apcprd04.prod.outlook.com>
Message-ID: <CALwnpeQDO499Ci0Td55D=o6L6kj0Wug3bpX=dt2QtNE-HAoBTA@mail.gmail.com>

Dear Fan,
To begin with, what is the sign of the B_Cost?


Regards,

Medard

On Thu, Aug 17, 2017 at 11:49 AM, FAN Zhou via Limdep <
limdep at mailman.sydney.edu.au> wrote:

> Hi all,
>
>
> I have run a Mixed logit model with Nlogit 6. But I am confused by the
> results.
>
>
> I was wondering how to interpret the insignificant mean but significant
> standard distribution for a random parameter. For example, if I set "cost"
> as random parameter, the mean 'B_COST' is insignificant but its associated
> standard deviation "NsCOST" is significant. Does results like this make
> sense?
>
>
> I appreciate any help.
>
>
> Cheers,
>
> Fan
> _______________________________________________
> Limdep site list
> Limdep at mailman.sydney.edu.au
> http://limdep.itls.usyd.edu.au
>
>

From zhoufan88 at hotmail.com  Thu Aug 17 19:13:15 2017
From: zhoufan88 at hotmail.com (FAN Zhou)
Date: Thu, 17 Aug 2017 09:13:15 +0000
Subject: [Limdep Nlogit List] Interpretation of Mixed logit model
In-Reply-To: <CALwnpeQDO499Ci0Td55D=o6L6kj0Wug3bpX=dt2QtNE-HAoBTA@mail.gmail.com>
References: <KL1PR04MB1221A2804597FB393C036709AE830@KL1PR04MB1221.apcprd04.prod.outlook.com>,
	<CALwnpeQDO499Ci0Td55D=o6L6kj0Wug3bpX=dt2QtNE-HAoBTA@mail.gmail.com>
Message-ID: <KL1PR04MB1221F117F141CBED1073D54FAE830@KL1PR04MB1221.apcprd04.prod.outlook.com>

Hi Medard,


Thank you very much for your reply.


Since I used log-normal distribution, and the 'cost' variable represent financial cost, the sign is forced to be negative.


Cheers,

Fan

________________________________
From: Limdep <limdep-bounces at mailman.sydney.edu.au> on behalf of medard kakuru via Limdep <limdep at mailman.sydney.edu.au>
Sent: Thursday, 17 August 2017 19:10
To: Limdep and Nlogit Mailing List
Cc: medard kakuru
Subject: Re: [Limdep Nlogit List] Interpretation of Mixed logit model

Dear Fan,
To begin with, what is the sign of the B_Cost?


Regards,

Medard

On Thu, Aug 17, 2017 at 11:49 AM, FAN Zhou via Limdep <
limdep at mailman.sydney.edu.au> wrote:

> Hi all,
>
>
> I have run a Mixed logit model with Nlogit 6. But I am confused by the
> results.
>
>
> I was wondering how to interpret the insignificant mean but significant
> standard distribution for a random parameter. For example, if I set "cost"
> as random parameter, the mean 'B_COST' is insignificant but its associated
> standard deviation "NsCOST" is significant. Does results like this make
> sense?
>
>
> I appreciate any help.
>
>
> Cheers,
>
> Fan
> _______________________________________________
> 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


From miq at wne.uw.edu.pl  Thu Aug 17 19:18:14 2017
From: miq at wne.uw.edu.pl (=?UTF-8?Q?Miko=c5=82aj_Czajkowski?=)
Date: Thu, 17 Aug 2017 11:18:14 +0200
Subject: [Limdep Nlogit List] Interpretation of Mixed logit model
In-Reply-To: <KL1PR04MB1221F117F141CBED1073D54FAE830@KL1PR04MB1221.apcprd04.prod.outlook.com>
References: <KL1PR04MB1221A2804597FB393C036709AE830@KL1PR04MB1221.apcprd04.prod.outlook.com>
	<CALwnpeQDO499Ci0Td55D=o6L6kj0Wug3bpX=dt2QtNE-HAoBTA@mail.gmail.com>
	<KL1PR04MB1221F117F141CBED1073D54FAE830@KL1PR04MB1221.apcprd04.prod.outlook.com>
Message-ID: <6cf8bf01-f912-462d-1be6-c43116aa5249@wne.uw.edu.pl>


Fan,

Note that for lognormally distributed parameters the estimates for the 
underlying normal are typically reported. So, even though the mean of 
the normal is = 0 the mean of lognormal would be positive and could be 
significant.

Cheers,
Mik


On 2017-08-17 11:13, FAN Zhou via Limdep wrote:
> Hi Medard,
>
>
> Thank you very much for your reply.
>
>
> Since I used log-normal distribution, and the 'cost' variable represent financial cost, the sign is forced to be negative.
>
>
> Cheers,
>
> Fan
>
> ________________________________
> From: Limdep <limdep-bounces at mailman.sydney.edu.au> on behalf of medard kakuru via Limdep <limdep at mailman.sydney.edu.au>
> Sent: Thursday, 17 August 2017 19:10
> To: Limdep and Nlogit Mailing List
> Cc: medard kakuru
> Subject: Re: [Limdep Nlogit List] Interpretation of Mixed logit model
>
> Dear Fan,
> To begin with, what is the sign of the B_Cost?
>
>
> Regards,
>
> Medard
>
> On Thu, Aug 17, 2017 at 11:49 AM, FAN Zhou via Limdep <
> limdep at mailman.sydney.edu.au> wrote:
>
>> Hi all,
>>
>>
>> I have run a Mixed logit model with Nlogit 6. But I am confused by the
>> results.
>>
>>
>> I was wondering how to interpret the insignificant mean but significant
>> standard distribution for a random parameter. For example, if I set "cost"
>> as random parameter, the mean 'B_COST' is insignificant but its associated
>> standard deviation "NsCOST" is significant. Does results like this make
>> sense?
>>
>>
>> I appreciate any help.
>>
>>
>> Cheers,
>>
>> Fan
>> _______________________________________________
>> 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
>


From zhoufan88 at hotmail.com  Thu Aug 17 19:19:09 2017
From: zhoufan88 at hotmail.com (FAN Zhou)
Date: Thu, 17 Aug 2017 09:19:09 +0000
Subject: [Limdep Nlogit List] Interpretation of Mixed logit model
In-Reply-To: <38232f4b-1321-3595-46e1-1b6fa7ff3035@wne.uw.edu.pl>
References: <KL1PR04MB1221A2804597FB393C036709AE830@KL1PR04MB1221.apcprd04.prod.outlook.com>,
	<38232f4b-1321-3595-46e1-1b6fa7ff3035@wne.uw.edu.pl>
Message-ID: <KL1PR04MB122168000482D90473260718AE830@KL1PR04MB1221.apcprd04.prod.outlook.com>

Dear Mik,


Thank you very much for your response.


I agree with your interpretation of such results. However, I wonder if it is worthwhile to further explore the heterogeneity in mean? Using the intersections, for example, ";rpl income". Because there is heterogeneity in presence, it seems the factors contribute to it should be tested, however, the mean itself is insignificant.

Cheers,
Fan

________________________________
From: Limdep <limdep-bounces at mailman.sydney.edu.au> on behalf of Miko?aj Czajkowski <miq at wne.uw.edu.pl>
Sent: Thursday, 17 August 2017 19:01
To: Limdep and Nlogit Mailing List
Subject: Re: [Limdep Nlogit List] Interpretation of Mixed logit model


Dear Fan,

in mixed logit a parameter is described using a parametric probability
distribution. This distribution could have a mean that is close to zero
(not significantly different) and some non-zero standard deviation
(significantly different from 0). This is a fine result meaning that on
average (mean) the effect is close to 0, but there is some heterogeneity
around this. On the other hand, if standard deviation is not
significantly different from 0 - you might as well model such a
parameter as fixed, as heterogeneity around the mean is not significant.

I cannot say if it makes sense for your variables / model, but
technically such a result is fine.

Best,
Mik



On 2017-08-17 10:49, FAN Zhou via Limdep wrote:
> Hi all,
>
>
> I have run a Mixed logit model with Nlogit 6. But I am confused by the results.
>
>
> I was wondering how to interpret the insignificant mean but significant standard distribution for a random parameter. For example, if I set "cost" as random parameter, the mean 'B_COST' is insignificant but its associated standard deviation "NsCOST" is significant. Does results like this make sense?
>
>
> I appreciate any help.
>
>
> Cheers,
>
> Fan
> _______________________________________________
> 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


From zhoufan88 at hotmail.com  Thu Aug 17 19:38:09 2017
From: zhoufan88 at hotmail.com (FAN Zhou)
Date: Thu, 17 Aug 2017 09:38:09 +0000
Subject: [Limdep Nlogit List] Interpretation of Mixed logit model
In-Reply-To: <6cf8bf01-f912-462d-1be6-c43116aa5249@wne.uw.edu.pl>
References: <KL1PR04MB1221A2804597FB393C036709AE830@KL1PR04MB1221.apcprd04.prod.outlook.com>
	<CALwnpeQDO499Ci0Td55D=o6L6kj0Wug3bpX=dt2QtNE-HAoBTA@mail.gmail.com>
	<KL1PR04MB1221F117F141CBED1073D54FAE830@KL1PR04MB1221.apcprd04.prod.outlook.com>,
	<6cf8bf01-f912-462d-1be6-c43116aa5249@wne.uw.edu.pl>
Message-ID: <KL1PR04MB12217B5A55D0121D93045AABAE830@KL1PR04MB1221.apcprd04.prod.outlook.com>

Hi Mik,


Thanks a lot for your response.


Cheers,

Fan


________________________________
From: Limdep <limdep-bounces at mailman.sydney.edu.au> on behalf of Miko?aj Czajkowski <miq at wne.uw.edu.pl>
Sent: Thursday, 17 August 2017 19:18
To: Limdep and Nlogit Mailing List
Subject: Re: [Limdep Nlogit List] Interpretation of Mixed logit model


Fan,

Note that for lognormally distributed parameters the estimates for the
underlying normal are typically reported. So, even though the mean of
the normal is = 0 the mean of lognormal would be positive and could be
significant.

Cheers,
Mik


On 2017-08-17 11:13, FAN Zhou via Limdep wrote:
> Hi Medard,
>
>
> Thank you very much for your reply.
>
>
> Since I used log-normal distribution, and the 'cost' variable represent financial cost, the sign is forced to be negative.
>
>
> Cheers,
>
> Fan
>
> ________________________________
> From: Limdep <limdep-bounces at mailman.sydney.edu.au> on behalf of medard kakuru via Limdep <limdep at mailman.sydney.edu.au>
> Sent: Thursday, 17 August 2017 19:10
> To: Limdep and Nlogit Mailing List
> Cc: medard kakuru
> Subject: Re: [Limdep Nlogit List] Interpretation of Mixed logit model
>
> Dear Fan,
> To begin with, what is the sign of the B_Cost?
>
>
> Regards,
>
> Medard
>
> On Thu, Aug 17, 2017 at 11:49 AM, FAN Zhou via Limdep <
> limdep at mailman.sydney.edu.au> wrote:
>
>> Hi all,
>>
>>
>> I have run a Mixed logit model with Nlogit 6. But I am confused by the
>> results.
>>
>>
>> I was wondering how to interpret the insignificant mean but significant
>> standard distribution for a random parameter. For example, if I set "cost"
>> as random parameter, the mean 'B_COST' is insignificant but its associated
>> standard deviation "NsCOST" is significant. Does results like this make
>> sense?
>>
>>
>> I appreciate any help.
>>
>>
>> Cheers,
>>
>> Fan
>> _______________________________________________
>> 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
>

_______________________________________________
Limdep site list
Limdep at mailman.sydney.edu.au
http://limdep.itls.usyd.edu.au


From zhoufan88 at hotmail.com  Thu Aug 17 19:40:49 2017
From: zhoufan88 at hotmail.com (FAN Zhou)
Date: Thu, 17 Aug 2017 09:40:49 +0000
Subject: [Limdep Nlogit List] Interpretation of Mixed logit model
In-Reply-To: <6cf8bf01-f912-462d-1be6-c43116aa5249@wne.uw.edu.pl>
References: <KL1PR04MB1221A2804597FB393C036709AE830@KL1PR04MB1221.apcprd04.prod.outlook.com>
	<CALwnpeQDO499Ci0Td55D=o6L6kj0Wug3bpX=dt2QtNE-HAoBTA@mail.gmail.com>
	<KL1PR04MB1221F117F141CBED1073D54FAE830@KL1PR04MB1221.apcprd04.prod.outlook.com>,
	<6cf8bf01-f912-462d-1be6-c43116aa5249@wne.uw.edu.pl>
Message-ID: <KL1PR04MB12214FFF2AE2DD2A51964B49AE830@KL1PR04MB1221.apcprd04.prod.outlook.com>

Hi Mik,


I think such result could happen when there is diversity in preference (e.g., positive and negative). Further exploration should be of interest, right?


Cheers,

Fan

________________________________
From: Limdep <limdep-bounces at mailman.sydney.edu.au> on behalf of Miko?aj Czajkowski <miq at wne.uw.edu.pl>
Sent: Thursday, 17 August 2017 19:18
To: Limdep and Nlogit Mailing List
Subject: Re: [Limdep Nlogit List] Interpretation of Mixed logit model


Fan,

Note that for lognormally distributed parameters the estimates for the
underlying normal are typically reported. So, even though the mean of
the normal is = 0 the mean of lognormal would be positive and could be
significant.

Cheers,
Mik


On 2017-08-17 11:13, FAN Zhou via Limdep wrote:
> Hi Medard,
>
>
> Thank you very much for your reply.
>
>
> Since I used log-normal distribution, and the 'cost' variable represent financial cost, the sign is forced to be negative.
>
>
> Cheers,
>
> Fan
>
> ________________________________
> From: Limdep <limdep-bounces at mailman.sydney.edu.au> on behalf of medard kakuru via Limdep <limdep at mailman.sydney.edu.au>
> Sent: Thursday, 17 August 2017 19:10
> To: Limdep and Nlogit Mailing List
> Cc: medard kakuru
> Subject: Re: [Limdep Nlogit List] Interpretation of Mixed logit model
>
> Dear Fan,
> To begin with, what is the sign of the B_Cost?
>
>
> Regards,
>
> Medard
>
> On Thu, Aug 17, 2017 at 11:49 AM, FAN Zhou via Limdep <
> limdep at mailman.sydney.edu.au> wrote:
>
>> Hi all,
>>
>>
>> I have run a Mixed logit model with Nlogit 6. But I am confused by the
>> results.
>>
>>
>> I was wondering how to interpret the insignificant mean but significant
>> standard distribution for a random parameter. For example, if I set "cost"
>> as random parameter, the mean 'B_COST' is insignificant but its associated
>> standard deviation "NsCOST" is significant. Does results like this make
>> sense?
>>
>>
>> I appreciate any help.
>>
>>
>> Cheers,
>>
>> Fan
>> _______________________________________________
>> 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
>

_______________________________________________
Limdep site list
Limdep at mailman.sydney.edu.au
http://limdep.itls.usyd.edu.au


From macario.ent at gmail.com  Wed Aug 23 09:32:41 2017
From: macario.ent at gmail.com (Mac RE)
Date: Wed, 23 Aug 2017 01:32:41 +0200
Subject: [Limdep Nlogit List] PDS for time-series data
In-Reply-To: <SY3PR01MB1097E3DC3EECCAEF40C894C7EB880@SY3PR01MB1097.ausprd01.prod.outlook.com>
References: <SY3PR01MB1097E3DC3EECCAEF40C894C7EB880@SY3PR01MB1097.ausprd01.prod.outlook.com>
Message-ID: <CADtfthMEBGTvDdH=vv1W1Nu2VXdFn1eE6L0C0aAN1UK2F6381A@mail.gmail.com>

Hi Yasir,
You can create a variable to specify the periods (T) individually and then
using this variable instead of a fixed number.
Kind regards,
Macario

2017-08-10 2:32 GMT+02:00 Yasir Ali via Limdep <limdep at mailman.sydney.edu.au
>:

> Hello everyone,
>
> I am using time-series data in which I have a response of participant at
> series of values. The time is not fixed, so I have a different set of
> values for one participant and similarly different for another participant.
> Is there any way to capture this repetition of values in MNL model?
> I know about PDS in which we define a number of repetition, in my case,
> repetition varies.
> Any help would be appreciated. Thanks.
>
> Regards,
> Yasir
> _______________________________________________
> Limdep site list
> Limdep at mailman.sydney.edu.au
> http://limdep.itls.usyd.edu.au
>
>