Sat 19 Nov, 2016 05:34 pm
I'm working on a discrete choice model, and I'd like to compute the elasticity of some of the parameters, such as the time. So from my discrete choice model, I have the estimation of the beta parameter for time.
Then, I can also compute the market share of a certain alternative.
With these two data, I'd like to use the formula (to use in the case of logit model with linear parameter) :
Elasticity = (1-P)*x*Beta
where P is the market share of the considered alternative, and Beta the estimated parameter of time.
However, in this formula, I don't know how to interpret x which is supposed to be the attribute (in this case the time). Should I just not consider it in the computation, or does it have to be a value ? Like the average time or something like this ?
Let me know if you want clearer explanation
How would you know market share? Do you know all the suppliers, and their share of the market?