As discussed on a previous thread I started on the Choice Experiments - General page ('A question about blocking'), I am designing my first DCE study, which has the following characteristics:
- 7 attributes, each with 5 levels
- 120 choice sets, split into 10 blocks -> 12 choices sets per respondent
- 1,000 participants, so 100 per block
- We plan to use MNL and MXL regression models to analyse the data, and will be interested in main effects only
- We have data from a previous pilot study (n=171), so are hoping to use a Bayesian efficient design to take into account these priors
My code (so far-please correct me if I've done anything wrong!) with no Bayesian priors is:
- Code: Select all
? Syntax for MNL Model (No Priors)
Design
;alts = ChoiceA*, ChoiceB*
;rows = 120
;block = 10
;eff = (mnl,d)
;model:
U(ChoiceA) = b0
+ b1_Tired * Tired[1,2,3,4,5]
+ b2_Walking * Walking[1,2,3,4,5]
+ b3_Sports * Sports[1,2,3,4,5]
+ b4_Concentration * Concentration[1,2,3,4,5]
+ b5_Embaressed * Embaressed[1,2,3,4,5]
+ b6_Unhappy * Unhappy[1,2,3,4,5]
+ b7_Treated * Treated[1,2,3,4,5] /
U(ChoiceB) =
b1_Tired * Tired
+ b2_Walking * Walking
+ b3_Sports * Sports
+ b4_Concentration * Concentration
+ b5_Embaressed * Embaressed
+ b6_Unhappy * Unhappy
+ b7_Treated * Treated
$
As the title of this post suggests, I'm now thinking about trying to incorporate the Bayesian priors from the pilot study into the experimental design.
However, the issue I've got is that some of the priors from the pilot study do not make intuitive sense. I've pasted the coefficients and standard errors from the MNL regression models (n=171) below.
- Code: Select all
Coeff Std Error
tired_2 0.097 0.109
tired_3 0.119 0.106
tired_4 -0.161 0.108
tired_5 -0.383 0.105
walki_2 0.064 0.106
walki_3 -0.202 0.107
walki_4 0.129 0.104
walki_5 -0.186 0.107
sport_2 0.248 0.112
sport_3 0.212 0.106
sport_4 0.428 0.110
sport_5 0.182 0.107
conce_2 -0.056 0.105
conce_3 0.153 0.105
conce_4 0.098 0.105
conce_5 -0.159 0.102
embar_2 0.260 0.110
embar_3 -0.065 0.109
embar_4 0.227 0.106
embar_5 0.134 0.105
unhap_2 0.077 0.107
unhap_3 0.090 0.106
unhap_4 -0.013 0.105
unhap_5 -0.212 0.106
treat_2 -0.074 0.109
treat_3 -0.181 0.107
treat_4 -0.215 0.108
treat_5 -0.530 0.106
Essentially, a priori one would expect ALL the coefficients to be negative, and for the coefficients to be increasing in magnitude within the attributes (i.e. the tired_1 and tired_5 coefficients should both be negative, and tired_5 should be larger in magnitude).
So my question is, if I were to use some of the priors from the pilot study (i.e. the ones which make intuitive sense) but not others, would this be a problem?
From looking at similar posts on the forum, I note that because I know the direction the coefficients should be going in, I could assign instead assign some of the coefficients a "very small negative or positive value, e.g. -0.000001 or 0.000001 as prior. This means that the prior is essentially zero, but it would allow the automatic avoidance of dominant alternatives in Ngene".
Any help regarding this matter would be greatly appreciated. Apologies if I haven't explained myself very well or have left out some information.
Best wishes,
Tom