Maximization function
We saw that the EM algorithm is an iterative method between two steps: the expectation and the maximization. In the last exercise, you created the expectation function. Now, create the maximization function which takes the data frame with the probabilities and outputs the estimations of the means and proportions.
This exercise is part of the course
Mixture Models in R
Exercise instructions
Create the function maximization by completing the sample code.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
maximization <- function(___){
means_estimates <- data_with_probs %>%
summarise(mean_1 = sum(x * ___) / ___(prob_cluster1),
mean_2 = sum(x * ___) / ___(prob_cluster2)) %>%
as.numeric()
props_estimates <- data_with_probs %>%
summarise(proportion_1 = ___(prob_cluster1),
proportion_2 = 1 - ___) %>%
as.numeric()
list(means_estimates, props_estimates)
}