Count data with consecutive counts
Posted: Wed Oct 12, 2016 5:27 am
Hi there,
I was wondering how to approach predicting the length an individual/observation is in a specific process.
Basically, I have a dataset which contains a positive number of days that an individual is in a process, which I am trying to predict.
This lead me fairly quick to modelling count data by means of a Poisson or Negative Binomial Regression, however, this only entails a number of times an event occurs within a specific time frame as far as I understand.
I also thought of approaching this problem as an ordinal classification problem or from a survival analysis perspective.
Does anyone have some thoughts on this subject?
Any help is appreciated!
I was wondering how to approach predicting the length an individual/observation is in a specific process.
Basically, I have a dataset which contains a positive number of days that an individual is in a process, which I am trying to predict.
This lead me fairly quick to modelling count data by means of a Poisson or Negative Binomial Regression, however, this only entails a number of times an event occurs within a specific time frame as far as I understand.
I also thought of approaching this problem as an ordinal classification problem or from a survival analysis perspective.
Does anyone have some thoughts on this subject?
Any help is appreciated!