I'm all in favor of better python interfaces for excel, but you need a better example, computing rolling time series statistics, and visualizing them? That's a

I'm all in favor of better python interfaces for excel, but you need a better example, computing rolling time series statistics, and visualizing them? That's a job for pandas. If you want excel as the front end (I understand why you would want that) you could use pandas to process the data, and use excel as the frontend, instead of writing all that code.