I am working on a problem wherein I am trying to predict how much 'percentage a student will score exams' based on the 'number of hours the student study', 'number of hours the student play', 'student's attendance in class', 'students hobbies', 'gender', 'his parents education' etc.
Now the challenge I am facing is that I am not sure which Generalized Linear Model or any other machine learning technique to use as the dependent variable i.e. 'percentage a student will score' will be a continuous value between 0 to 100 which means the prediction should be non-negative and continuous value between 0 to 100. I have plotted the histogram and the distribution looks a little left skewed.
I tried using Linear regression but it is predicting negative values and also values more than 100. I tried Negative binomial but even that is not working as it is predicting only discreet values. Please let me know which GLM technique will solve this issue.