Advice on which logit model to use

[Hannah Sack]

Dear All

I have conducted a choice experiment on eco - labels and now need to estimate a logit model to estimate utilities for the specific attributes and attribute levels.

I have four attributes each with different levels for each attribute. Each respondent was given four options to choose from with the fourth option given as a none option. Each respondent completed 8 choice tasks. All the attributes for the choice tasks were randomly interacted so it is not the case that all the choice tasks are the same for all individuals.

I am now having trouble deciding which logit model to use: clogit asclogit or mixed logit.

Below I have inserted a picture of what my data looks like. I have tried to run the code asclogit Choice i.Countryoforigin i.HarvestMethod i.SustainabilityLabel Price, case(Case) alternative(Option) vce (cluster Person) basealternative(4) however this does not work.

Can someone please give me some advice on how to proceed. Many thanks

[Richard Williams]

Hannah, it would be better to present your data using dataex. Then someone who wanted to could try out different analyses. See pt 12 of the FAQ

[Hannah Sack]

Dear Richard. Thank you. Please find my data below:

* Example generated by -dataex-. To install: ssc install dataex
clear
input int Person byte Task float(Option Case) byte(Choice Countryoforigin HarvestMethod SustainabilityLabel Price)
1 1 1  1 0 2 2 2 4
1 1 2  1 0 3 2 3 2
1 1 3  1 1 1 1 1 1
1 1 4  1 0 0 0 0 0
1 2 1  2 1 3 2 1 5
1 2 2  2 0 1 1 2 6
1 2 3  2 0 2 1 3 3
1 2 4  2 0 0 0 0 0
1 3 1  3 0 2 1 2 2
1 3 2  3 1 3 1 1 4
1 3 3  3 0 1 2 3 1
1 3 4  3 0 0 0 0 0
1 4 1  4 1 2 2 1 6
1 4 2  4 0 1 2 3 3
1 4 3  4 0 3 1 2 5
1 4 4  4 0 0 0 0 0
1 5 1  5 1 1 2 1 2
1 5 2  5 0 3 1 3 6
1 5 3  5 0 2 2 2 1
1 5 4  5 0 0 0 0 0
1 6 1  6 0 2 1 3 5
1 6 2  6 1 1 1 1 4
1 6 3  6 0 3 2 2 3
1 6 4  6 0 0 0 0 0
1 7 1  7 0 1 2 2 5
1 7 2  7 0 3 1 3 1
1 7 3  7 1 2 1 1 3
1 7 4  7 0 0 0 0 0
1 8 1  8 0 2 2 3 4
1 8 2  8 0 1 1 2 2
1 8 3  8 1 3 2 1 6
1 8 4  8 0 0 0 0 0
3 1 1  9 0 2 1 3 2
3 1 2  9 1 1 1 1 1
3 1 3  9 0 3 2 2 4
3 1 4  9 0 0 0 0 0
3 2 1 10 0 2 2 1 5
3 2 2 10 0 1 1 2 6
3 2 3 10 1 3 2 3 3
3 2 4 10 0 0 0 0 0
3 3 1 11 0 1 2 2 5
3 3 2 11 1 3 1 1 1
3 3 3 11 0 2 2 3 6
3 3 4 11 0 0 0 0 0
3 4 1 12 0 3 1 3 4
3 4 2 12 1 2 2 2 2
3 4 3 12 0 1 1 1 3
3 4 4 12 0 0 0 0 0
3 5 1 13 1 1 2 3 1
3 5 2 13 0 2 1 1 5
3 5 3 13 0 3 1 2 6
3 5 4 13 0 0 0 0 0
3 6 1 14 1 3 2 1 2
3 6 2 14 0 1 2 2 4
3 6 3 14 0 2 1 3 3
3 6 4 14 0 0 0 0 0
3 7 1 15 0 3 2 3 5
3 7 2 15 1 2 1 2 1
3 7 3 15 0 1 2 1 6
3 7 4 15 0 0 0 0 0
3 8 1 16 1 3 1 3 2
3 8 2 16 0 2 1 1 4
3 8 3 16 0 1 2 2 3
3 8 4 16 0 0 0 0 0
4 1 1 17 0 1 1 3 4
4 1 2 17 1 2 2 1 2
4 1 3 17 0 3 1 2 3
4 1 4 17 0 0 0 0 0
4 2 1 18 0 1 1 1 1
4 2 2 18 0 3 2 2 6
4 2 3 18 0 2 2 3 5
4 2 4 18 1 0 0 0 0
4 3 1 19 0 2 2 2 1
4 3 2 19 0 1 2 3 3
4 3 3 19 0 3 1 1 5
4 3 4 19 1 0 0 0 0
4 4 1 20 0 3 2 1 4
4 4 2 20 0 1 1 3 6
4 4 3 20 0 2 1 2 2
4 4 4 20 1 0 0 0 0
4 5 1 21 0 3 1 3 2
4 5 2 21 0 1 2 2 5
4 5 3 21 1 2 2 1 6
4 5 4 21 0 0 0 0 0
4 6 1 22 0 2 1 2 4
4 6 2 22 0 1 1 1 3
4 6 3 22 0 3 2 3 1
4 6 4 22 1 0 0 0 0
4 7 1 23 0 3 1 2 6
4 7 2 23 0 2 2 3 3
4 7 3 23 0 1 2 1 2
4 7 4 23 1 0 0 0 0
4 8 1 24 0 2 1 3 1
4 8 2 24 0 3 2 2 4
4 8 3 24 0 1 1 1 5
4 8 4 24 1 0 0 0 0
5 1 1 25 0 1 2 3 6
5 1 2 25 1 2 1 1 4
5 1 3 25 0 3 1 2 3
5 1 4 25 0 0 0 0 0
end
label values Case Case
label def Case 1 "1 1", modify
label def Case 2 "1 2", modify
label def Case 3 "1 3", modify
label def Case 4 "1 4", modify
label def Case 5 "1 5", modify
label def Case 6 "1 6", modify
label def Case 7 "1 7", modify
label def Case 8 "1 8", modify
label def Case 9 "3 1", modify
label def Case 10 "3 2", modify
label def Case 11 "3 3", modify
label def Case 12 "3 4", modify
label def Case 13 "3 5", modify
label def Case 14 "3 6", modify
label def Case 15 "3 7", modify
label def Case 16 "3 8", modify
label def Case 17 "4 1", modify
label def Case 18 "4 2", modify
label def Case 19 "4 3", modify
label def Case 20 "4 4", modify
label def Case 21 "4 5", modify
label def Case 22 "4 6", modify
label def Case 23 "4 7", modify
label def Case 24 "4 8", modify
label def Case 25 "5 1", modify

[Phil Bromiley]

The answer to your question requires more information. Are the four choices ordered - suggesting an ordinal logit or probit (ologit or oprobit, or xtologit with person as panel), or are they just categories (multinomial logit or probit)? I'm not sure if there is a panel variant of mlogit or mprobit. Are you trying to control for an individual effect or what?

[Richard Williams]

For random effects mlogit, see

hhttps://www.stata-journal.com/articl...article=st0362

For fixed effects mlogit, see

https://www.stata-journal.com/articl...article=st0362

Answers to Phil’s questions may be helpful in choosing a strategy.