MCDEV for land allocation

[Maksym]

Dear All,

I would like to use the MCDEV model to model land allocation for tree planting on farms. They can choose areas of none, one or several (out of 11) types of tree planting. The total area is small relative to the remaining area (I assume an “outside good”). Is this an appropriate model?

The examples use the same alpha for all alternatives. Would it make sense to use alternative-specific alphas? Should then gammas be fixed?

Thank you.

[stephanehess]

Hi

the decision on which profile to use (generic alpha, etc) is an empirical question. I suggest you have a look at Bhat's initial paper on this.

Regarding an outside good, this depends. If you're only looking at a specific area, and if the remaining area is fixed, then there is no need to include that

Stephane

[Maksym]

Thank you, Stephane, I will try to compare different profiles.

Is the MCDEV appropriate for the choices I try to model? I would like to clarify that the total area (budget) varies and ranges from ~10 to ~1000 ha, and the sum of areas of selected activities ranges from 0 to 30% of the total area. The remaining area is not fixed, so my understanding it needs to be outside good.

A follow-up question is about predicted continuous consumption apollo_prediction(). Are these conditional on an alternative being selected?

Thank you.

Maksym

[stephanehess]

Hi

I'm not sure I understand your data. Can you explain it a bit better, maybe with some examples from the data?

Regarding prediction, please consult the manual. It is the expected consumption at the observation level

Stephane

[Maksym]

Dear Stephane,

This is a choice experiment where farmers are presented with hypothetical incentive programmes and choose whether they would plant different types of tree plantings and what areas would they plant.

Below is a snippet of a simulated simple dataset.

Variables:

• "ID" and "Obs" are respondent and response.
• "Size" is the area of a farm ("budget").
• "a_cp" to a_sb are areas of different types of plantings that may be selected to plant ("goods").
• "a_ou" is the remaining area ("outside good").
• "grant" is the payment, $000/ha, "fill" is help with the application, "seedl" is the provision of seedlings (characteristics of a programme).
• "steep" is the topography of a farm (characteristic of the respondent)

Is the MCDEV appropriate for the choices I try to model?

Thank you,

Maksym

Code: Select all

ID,Obs,Size,a_cp,a_cn,a_cb,a_bp,a_bn,a_bb,a_ln,a_lp,a_lb,a_sn,a_sb,a_ou,grant,fill,seedl,steep
1,1,1000,0,0,0,0,0,0,0,0,0,0,0,1000,1,0,1,0.75
1,2,1000,0,0,0,0,0,0,0,0,0,0,0,1000,0,1,0,0.75
1,3,1000,0,0,0,0,0,0,0,50,0,0,0,950,6,0,1,0.75
2,1,31,0,0,0,0,0,0,0,0,0,0,0,31,0,1,0,0.5
2,2,31,0,0,1.2,0,0,0,0,0,0,0,0.8,29,6,1,1,0.5
2,3,31,0,0,0,0,0,0,0,0,0,0,0,31,6,0,0,0.5
3,1,43,0,0,0,0,0,0,0,0,0,0,0,43,6,1,1,0.75
3,2,43,0,0,0,0,0,0,0,0,0,0,0,43,1,1,0,0.75
3,3,43,0,0,0,0,0,0,0,0,0,0,0,43,0,1,1,0.75
4,1,100,0,0,0,0,0,0,0,0,0,0,0,100,1,0,1,0.25
4,2,100,0,0,0,0,0,0,0,0,0,0,0,100,0,0,0,0.25
4,3,100,0,0,0,0,0,0,0,0,0,0,0,100,12,0,1,0.25
5,1,40,0,1,0,0,0,0,0,0,0,1,0,38,3,0,1,0.75
5,2,40,0,0,0,0,0,0,0,0,0,0,0,40,0,1,0,0.75
5,3,40,0,1,0,0,0,0,0,0,0,1,0,38,12,0,1,0.75
6,1,473,0,0,0,0,0,0,0,0,0,0,0,473,0,1,0,0.75
6,2,473,0,0,0,0,10,0,0,0,0,0,0,463,6,1,1,0.75
6,3,473,0,0,0,0,0,0,0,0,0,0,0,473,6,0,0,0.75
7,1,46,0,4,0,0,0,0,0,0,0,0,0,42,3,0,1,0
7,2,46,0,0,0,0,0,0,0,0,0,0,0,46,0,1,0,0
7,3,46,0,4,0,0,0,0,0,0,0,0,0,42,12,0,1,0
8,1,10,0,1.5,0,0,0,0,0.5,0,0,0,0,8,1,0,1,0.5
8,2,10,0,1.5,0,0,0,0,0.5,0,0,0,0,8,6,1,0,0.5
8,3,10,0,1.5,0,0,0,0,0.5,0,0,0,0,8,1,0,1,0.5
9,1,100,0,0,0,0,0,0,0,0,0,0,0,100,6,1,1,0.5
9,2,100,0,0,0,0,0,0,0,0,0,0,0,100,1,1,0,0.5
9,3,100,0,0,0,0,0,0,0,0,0,0,0,100,0,1,1,0.5
10,1,22,0,0,0,0,0,0,0,0,0,0,0,22,6,1,1,0
10,2,22,0,0,0,0,0,0,0,0,0,0,0,22,1,1,0,0
10,3,22,0,0,0,0,0,0,0,0,0,0,0,22,0,1,1,0

[stephanehess]

Hi

many apologies for the slow reply.

Yes, it looks like you could use MDCEV, defining the budget to be the area and including an outside good. But that will only work if there is never a case where the entire area is used up.

Stephane