3 weeks ago
I have a bunch of input variables (call center data, like number of calls, agent productivity, that sort of thing), and I need to predict one particular variable into the future. I have daily data that goes back for 11 months, and I'd like to make a prediction of the dependent variable over the next five years.
I had been using the ARIMA task, but it looks like I'm not able to input multiple independent variables in EG. Is an autoregressive model better?
Basically I'd like two pieces of output:
- given a future value of all of my input variables, what will my output value be?
- example: call volume is X, agent productivity is Y, therefore my output value will be some function of X and Y
- what is the confidence with which I know this?
Any help would be AWESOME! Thanks