|
I am trying to model the time lapse between when a user sees an ad and when they call the advertiser (presuming they do). I have two issues:
Is there a simple way to model this? |
|
How about an exGaussian distribution? It is the difference between a normal distribution and an exponential distribution and people use it to model reaction times in psychophysics. |
|
It is probably overkill for your application, but there is an excellent book on reaction times. It's a rare careful and thorough work through of reaction times, and the kind of things you can do with them (including the statistical approaches). I came across it after proposing to some psycho-linguist friends the idea of using hazard functions to model how long you'll wait to see a response. |
|
I heard about doubly-exponential or ex-gaussian distribution for RTs data (especially for spotting outliers), but you might have a look at the following papers (all available on the web):
With regard to your original question, I agree with Alexandre that it is worth to look at a Gamma distribution. |
