I have over 2700 customers that I need to analyze. It shows for each customer when in the last 5 years they made a purchase, based on months. It looks something like 0, 0, 100, 0, 0, 0, 0, 200, 0, 0, 0, 500 for a single year for a single customer. I have a formula that get produce the sizes of the gaps between each purchase. From the gap sizes I was able to find the mean gap size and the standard deviation of gap sizes. I then compared the time since the last purchase to find the standard deviations away from the mean gap size it is. This turned out to be a problem for a lot of my customers since this will not work if there is less than 3 purchases and if every gap between purchases is the same. I've tried quite a few methods to predict this data and cannot find any way to do so. If anyone can find a better way to do so please let me know
