I am using a simple implementation of Bayes theorem to find the discrete probability distribution of proportion of wins.

naiveBayes <- function(theta,prior,win=T)
      ifelse(win==T,likelihood <- theta,likelihood <- 1-theta)
      constant <- sum(prior * likelihood)
      posterior <- (prior * likelihood) / constant

A win being True or FALSE gives a posterior which I will use a prior for the second computation and so on.

Question: How to give more weight to recent event? Say, the new event is a loss (win=F), I would like it to have higher influence than what the bayes theorem provides.


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.