Want to get involved? We're always looking for ideas and content for Weekly Challenges.
SUBMIT YOUR IDEADoing a comparison of the time series models--(ETS and ARIMA), it looks like the ARIMA model follows the data very closely, but it fails to properly predict the huge spikes in the flu season; I would attribute this because the data does not really hold the criteria of a time series dataset, since there are many missing weeks without a value that would need to be imputed with zero. On the other hand, the ETS model does a much worse job projecting the following 52 weeks.