Estimating related metrics using Maximum A Posteriori
English is not my mother tongue; please excuse any errors on my part.
I've recently faced a problem for which I haven't found any solutions after investing a lot of time. Here is the summarized problem statement:
Imagine our dataset being the history of matches of a football team, and each point in our dataset is either zero (lost) or one (won). Here we define four metrics that we want to estimate;
- winning ratio (# of wins / # total matches)
- the index of the first win in the dataset
- the average distance of win indices in the dataset
- has the team won any of the first three matches?
I already know that the first metric is estimated using MLE or MAP, but I don't know how to estimate the other ones.
Topic estimation bayesian
Category Data Science