So now in this video we will be doing the prediction of our dependent variable that is customer satisfaction for our validation data and we will find the correlation between the observed value and the predicted value of customer satisfaction or dependent variable for our validation data to check whether the prediction is correct or not, or how close is our observed value and predicted value of our dependent variable for the validation data. So now let's start data validation underscore result. Set validation validation is very original data set, which contains 30% of my 30% of my original data set that is linear underscore underscore retail. So, this is the second part of the data lake I have divided my original data into two parts that is 70% 30% is 30% of the data that is validation. So validation data set is getting copied into validation underscore result. Now, I will be predicting the value of the customer satisfaction.

As you remember that we had received we had God word parameter estimate table in our last results, showing the coefficients of all the significant independent variables so we will be copying the variable names with the parameter estimates or coefficients in this code, and we will be crediting the value of customer satisfaction for our validation data. So predicted equals two. First we'll be copying the intercept term from here So, my intercept is minus 2.12140 class here we are basically forming the equation. Class. My next independent variable is product quality the same copy Here I'm copying all the coefficients with the significant variables I'm copying all the significant variables with efficients Then I've given the run statement and then quit. Now let me explain the code see data validation underscores are this is a data set, which I'm creating inside work as I did not specify any library to create inside work.

Validation is 30% of my data which is getting copied in validation underscore result. Now I'm creating I want to predict the value of me customer satisfaction. So as we know the standard form of linear regression equation is y is equal to alpha plus beta one x one plus beta two x two plus beta three x three, two plus b 10 x 10 plus EA. So here this is my intercept term plus These are my independent variables with the coefficients that is either a beta one beta two beta three beta four dot will be done. And now using these independent variables that is product quality, ecommerce advertising product line sales force image competitive pricing, packaging order billing price flexibility, I'm going to predict the value of my independent variable customer satisfaction for my validation data. So, let me run this code, I will get the variable that is the predicted value of customer satisfaction inside the data set validation underscore self that will be created in that work.

So, let me open this data set. So, this is the predicted value of customer satisfaction now, and find the correlation between the observed and predicted value for the validation data of our dependent variable customer satisfaction. So, I'm going to do crock-pot data validation underscore result customer satisfaction says me observed value that is observed variable predicted is a predicted value of a customer satisfaction, I want to find the correlation between these two to find how close are my observed and predicted values. So, let's find the correlation to see the 0.87. So, my coalition is quite good formal training data and correlation was 0.90. And for mine validation data is 0.87.

So which is quite good that means we predicted levels of customer satisfaction is quite close to observable. So my prediction is correct. So we will be learning till here in this video before Move to the next video. let me recap the concepts that we have done in now. Thank you Goodbye. see you for the next video.