Sports are more fun, but I have to admit they are less consequential in real life. For example, you might want to know the number of expected mutations due to radiation on a fixed stretch of DNA, or the number of plane crashes from mechanical malfunction in a year. Of course beyond props there are many applications of these distributions. Knowing the entire distribution is important, because props never ask you for point estimates. In this case, it means I know how likely it is that a player throws for 2 touchdowns, or 3, or 4, or 10 rather than just an average. The largest benefit is that inference methods, by definition, estimate the underlying distribution of the target variable. I’m excited to show off some inference methods to estimate touchdown props because the methods themselves are very powerful and have diverse application. Statistical framework for modeling discrete events with Poisson distributions Motivation
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |