I am trying to build a model (most likely a regression or random forest regression) for quarterly financial data. My training data has a daily cadence, but I am not sure how to work with these to predict a quarterly dependent variable. Clearly I could summarize these by quarter (avg, mean, etc.) but that would result in few total observations, and it feels like that is throwing out some valuable variance to train on.
Thanks for the help!
One way is to transform the daily data into dynamic quarterly data, eg by averaging over a moving window (that represents a quarter) and similarly for other statistics.
These windows can be overlapping, eg by a factor $o$%, no need to be non-overlapping.
This way the data are transformed to quarter period statistics while not actually reducing amount of data (but simply transformed).
Then you feed your regression model these quarterly data instead of the daily data in order to get quarterly results.
It is clear that to get results over quarter periods one needs inputs that represent quarter periods in the first place. So this is a natural way to achieve it.
