The aim is to find who will win the cricket World Cup'19 based on ODI match results from last 10 years. I have attached a figure of how the data looks now.
All the variables are nominal and have more than 2 classes. For now, I have coded them as dummy variables and have performed an SLR to get a basic idea of how the result will turn out to b
It seems multinomial logistic regression when taking in to account the dependent variable.
Asrat Chernet I tried multinomial regression, the model became way too complex. So, I have changed the dependent variable now to binary form and implemented binomial logistic regression.
@Raman you are not modeling the situation correctly and throwing away information is rarely a good idea. I am attaching an excellent set of notes by a real expert which I think should help you. Suggest you consult them carefully. Should you desire to try multinomial regression again, Google is a good source of advice. Good luck, David Booth
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression