Prediction of 3 output variables

Hi All,

I am trying to build a model which can predict the 3 output variables based on some independent variables.
So far I learnt how a model can predict single contineous variable using regression techniques and how to classify the binary or multi-lable variable.
But I am trying to find out a good appraoch to classify 3 variables.
For example, let say I have a loan data of different types of customer and the usual customer behaviour on paying EMI is
1.paid before the due date
2.paid after the due date
3.paid on time
I want to build a model that will predict my customer EMI payment whether it will be ON, Before or After the due date.
Thanks in advance for your help. Please see the uploaded image for sample data.


Hi Everyone,
i need to build a model which predict whether the student gonna attend next class or not based on previous class so please tell me how to choose X and Y variable?
Reg_no class1 class2 class3
BCOM01 1 1
bcom02 0 1
here 0 represents absent and 1 represent present
class3 i need to predict, that is my prediction variable
please help me out guys

You can apply naive Bayes algorithm


This kind of model you want to build is a “multi label” class problem.
It is not so common like the regresion, binary o multiclass, there is a scare number of literature about them.

Carlos (Spain)

© Copyright 2013-2019 Analytics Vidhya