choice-metrics.com

Hi,

I just created a model averaging design but needed to change some attribute levels manually. Then, I wanted to evaluate the design with the following syntax in Ngene:

design

;alts(m1) = ACC*, DEC*
;alts(m2) = ACC*, DEC*


;rows=24
;block=6
;eff = m1(mnl,d) + m2(mnl,d)

;eval=H:\My Documents ...

My decision is to set all priors to zero. Previously, you mentioned that even if I use no priors, ASC0 must have a value. Should I use those insignificant priors for ASC0? For instance Bayesian priors, for M1: ASC0[(n,-0.259,1.24)] and for M2: ASC0[(n,2.18,2.36)].

Here are the choice probability for both models:


MNL probabilities (m2)
Choice acc dec
1 0.998127 0.001873
2 0.932267 0.067733
3 0.99863 0.00137
4 0.935535 0.064465
5 0.761722 0.238278
6 0.739696 0.260304
7 0.732078 0.267922
8 0.992198 0.007802
9 0.466139 0.533861
10 0.727582 0.272418
11 0 ...

Thanks Michiel for your detailed response. Just one more point about ASC0. For M1, ACS0 = -0.259 (p-value=0.83) and for M2, ASC0=2.18 (p-value=0.36). Both of them are insignificant. If we include them, the utility of DEC is positive in M2 which makes sense but it is negative for M1 which seems to be ...

Brilliant! I applied this method to my survey which is about the taxi drivers' ride acceptance behaviour. The idea is that respondents (taxi drivers) receive a ride request in their mobile application and they just need to decide whether to accept or decline it (a binary choice) based two sets of ...

Thanks Michiel for your reply. Just ran the example design but the problem is that this model does not keep the same attribute levels for the common attributes (e.g., b1 and b2). We do not want to change the attribute levels of the first choice set given that the idea is to understand whether the ...

Hi Michiel,

First of all, I would like to appreciate your massive support. My question is about a particular design with two stages. At the first stage, respondents are supposed to choose an alternative based on the first set of attributes. Then, more attributes (the second sets of attributes) are ...

Thanks Michiel for your advice. I was wondering why you suggested -0.3 as the mean? This looks to be an extreme situation (-0.0866-0.210=-0.2966), right?
I also have another question. I am going to replace some attributes with some new ones, so the utility function is going to change. Can I still ...

Thanks Micheil for your help. As shown below, in some cases, the standard error is much larger than the estimated parameters:

Name Value Robust SE
B_time -0.0866 0.210
B_surge 0.0354. 0.215
B_wait 0.0379 0.0258

I was wondering if I can still use these priors for a Bayesian design?
If so, for the ...

Hi Michiel,

In my experiment, there are 2 alternatives and 8 contextual variables. I have conducted a pilot with 50 respondents (400 observations) to estimate the priors, but none of the parameters is significant and surprisingly some of them have unexpected (implausible) signs. I was wondering ...