Important: Read this before posting to this forum

  1. This forum is for questions related to the use of Apollo. We will answer some general choice modelling questions too, where appropriate, and time permitting. We cannot answer questions about how to estimate choice models with other software packages.
  2. There is a very detailed manual for Apollo available at http://www.ApolloChoiceModelling.com/manual.html. This contains detailed descriptions of the various Apollo functions, and numerous examples are available at http://www.ApolloChoiceModelling.com/examples.html. In addition, help files are available for all functions, using e.g. ?apollo_mnl
  3. Before asking a question on the forum, users are kindly requested to follow these steps:
    1. Check that the same issue has not already been addressed in the forum - there is a search tool.
    2. Ensure that the correct syntax has been used. For any function, detailed instructions are available directly in Apollo, e.g. by using ?apollo_mnl for apollo_mnl
    3. Check the frequently asked questions section on the Apollo website, which discusses some common issues/failures. Please see http://www.apollochoicemodelling.com/faq.html
    4. Make sure that R is using the latest official release of Apollo.
  4. If the above steps do not resolve the issue, then users should follow these steps when posting a question:
    1. provide full details on the issue, including the entire code and output, including any error messages
    2. posts will not immediately appear on the forum, but will be checked by a moderator first. This may take a day or two at busy times. There is no need to submit the post multiple times.

Unlabeled experiment (MNL model with SP Covariates)

Ask questions about model specifications. Ideally include a mathematical explanation of your proposed model.
Post Reply
Yashin
Posts: 21
Joined: 23 Sep 2021, 01:43

Unlabeled experiment (MNL model with SP Covariates)

Post by Yashin »

Dear Sir,

Q1. I am estimating an MNL model with SP covariates, generic alternatives, and only 4 attributes. Including 'ASC' in the utility function will help me to determine which of the alternative has been chosen/preferred to be chosen by the respondents mostly? While estimating, I see a small information on the output screen:

Overview of choices for MNL model component :
Option_1 Option_2 Option_3
Times available 2976.00 2976.00 2976.0
Times chosen 886.00 1087.00 1003.0
Percentage chosen overall 29.77 36.53 33.7
Percentage chosen when available 29.77 36.53 33.7

Can I conclude by saying that Option 2 is the most preferred option by the respondents?

Q2. While including the socio-demographics and other explanatory variables, the only way to add a meaningfully interpretable model is to have the interactions for all the utility functions since the model is restricted and unlabelled. I would like to make sure that,
if I am trying to capture the information of the non EV owner (in my case EVNO with Ecar_d=0 are the non-Ev owners) having an effect on distance and charging speed or not, and the variable ' KM' (kilometer) interacts only with the price coefficient to understand the sensitivity of the price with KM (kilometer driven), the stated utility functions (script below) mentioned below is the right approach to do so?


V[['Option_1']] = b_dist * Distance_1 +b_resv * Reservation_1 + b_cost * Price_1 +
b_cpeed * Charging_Speed_1 +b_EVNOD*((Ecar_d==0)*Distance_1)+
b_EVNOS*((Ecar_d==0)*Charging_Speed_1)+b_KMP*((KM)*Price_1)

V[['Option_2']] = b_dist * Distance_2 +b_resv * Reservation_2 + b_cost * Price_2 +
b_cpeed * Charging_Speed_2 +b_EVNOD*((Ecar_d==0)*Distance_2)+
b_EVNOS*((Ecar_d==0)*Charging_Speed_2)+b_KMP*((KM)*Price_2)

V[['Option_3']] = b_dist * Distance_3 +b_resv * Reservation_3 + b_cost * Price_3 +
b_cpeed * Charging_Speed_3 +b_EVNOD*((Ecar_d==0)*Distance_3)+
b_EVNOS*((Ecar_d==0)*Charging_Speed_3)+b_KMP*((KM)*Price_3)
stephanehess
Site Admin
Posts: 974
Joined: 24 Apr 2020, 16:29

Re: Unlabeled experiment (MNL model with SP Covariates)

Post by stephanehess »

Yashin

1. you should always include alternative specific constants as the model makes the assumption that the mean of the error terms is the same, so you need the constants to capture the mean impact of all factors not included in the deterministic utility function. Yes, the output shows you that option 2 is chosen the most often, but that can of course be due to both observed and unobserved factors.\

2. yes, those interactions should work, but you should consider whether you want them to be linear or not

Stephane
--------------------------------
Stephane Hess
www.stephanehess.me.uk
Post Reply