I am currently calculating a bayesian average using Q1 data using the following equation:

enter image description here

I would like to incorporate data from other quarters for several years back, i.e. Q4 2021, Q3 2021, Q2 2021, etc. I plan to use exponential decay to add weights to each quarter so that more recent quarters are weighted heavier. How would I go about creating the following variables: productRatingsAvg x productRatingsCount, and productRatingsCount in the denominator?

Would productRatingsAvg x productRatingsCount look like something as follows? If so, what do I use for the denominator? Total products from all the quarters or averaged?

[(Q1productRatingsCount x Q1productRatingsAvg)*weight1 + (Q2productRatingsCount x Q2productRatingsAvg)*weight2 + (Q3productRatingsCount x Q3productRatingsAvg)*weight3 + (Q4productRatingsCount x Q4productRatingsAvg)*weight4] / 4

Would C and m then be calculated over the total dataset? Or just the current quarter since it's the one I want weighted heaviest anyway?

Apologies for the formatting with my equation.



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.