2
$\begingroup$

I am given a time series data vector (ordered by months and years),which contains only 0s and 1s. 1 s represent a person changes his job at a particular a month.

Questions: What model can i use to determine model how frequently this person change his job ? In addition, this model should be able to predict the probability of this person changing his in the next 6 months.

A poisson process ? (I have studied poisson process before however I have no idea when and how to apply it). Any assumptions that data need to meet before applying the poisson process ?

Would love to gather more information on how to model something like this. Thanks

$\endgroup$
2
$\begingroup$

It would be reasonable to assume that each job duration is independent, so a renewal process might be a better model than a Poisson process. (Poisson process is a special case of renewal when the interval distribution is exponential.) Of course the next question is which interval distribution to use. Inverse Gaussian is a possibility.

EDIT: If you have plenty of data, you could make a mild assumption to justify some form of normalization of your intervals. For example, you could scale them to have the same mean, and look at the population distribution of the normalized interval to get an idea of which interval distribution would be appropriate.

$\endgroup$
  • $\begingroup$ Yes. I agree with what you said. But I kept asking myself which interval distribution to use ? And from the data I have, 29 12 24 14 27 9 17 58 is the list of number of month between events. (Events = this person changes his job). To be honest, you can pick "exponential", "normal", "inv Gamma".... to be the interval distribution. So my question is: Based on what criteria do we choose the interval distribution? $\endgroup$ – user1769197 Jul 17 '14 at 16:42
  • $\begingroup$ Is that all the data you have? I would gather more data from a larger population and do some nonparametric/histogram-like estimation to get a feeling, but if that's all you have, I don't know what you can do. $\endgroup$ – Memming Jul 17 '14 at 17:08
  • $\begingroup$ Thats the data for one particular person. $\endgroup$ – user1769197 Jul 17 '14 at 17:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.