You can implement a number of approaches: ARIMAResults include aic and bic. By their definition, (see here and here), these criteria penalize for the number of parameters in the model. So you may use these numbers to compare the models. Also scipy has optimize.brute which does grid search on the specified parameters space. So a … Read more
I would begin by visualizing my data to get some intuitions , detect patterns and choose the best model. R is really great for this. Here I am using ggplot2 but you can do the same thing using lattice. For example : dat$DATE <- as.Date(dat$DATE,format=”%d/%m/%Y”) library(ggplot2) ggplot(dat,aes(x=DATE,y=Rev,group=ProdCode,color=ProdCode)) + geom_line() + facet_grid(~ID) + geom_smooth(method=’lm’) Here dat … Read more