Conditional Logit model in R
Budget: $30 – $250 AUD
Produce a script to generate a conditional Multinomial Logistic Regression model in R (using mlogit, nnet, or other package) to predict the “Win” likelihood of horses based on the training set provided (odds_train_MLR.csv).
Code is required to build and optimise the model, and then apply the model to the test set provided (odds_test_MLR.csv).
The output will be a set of predictions for the upcoming races (between 0 and 1). The sum of the predictions for each race (Race_no) must equal one, as there is a 100% chance that one of the horses wins the race.
Code is required to build and optimise the model, and then apply the model to the test set provided (odds_test_MLR.csv).
The output will be a set of predictions for the upcoming races (between 0 and 1). The sum of the predictions for each race (Race_no) must equal one, as there is a 100% chance that one of the horses wins the race.