So in this video we will be using this independent variables to predict the value of our dependent variable for training data and we'll be finding the correlation between the observed value and the predicted value of customer satisfaction for our training data. So let's do it. So we are going to use the procedure proc reg. Data equals training model is my key word. Customer satisfaction is the dependent variable. product quality.

Now I have to copy this itself in dependent variables. So these are the independent variables or these are the significant variables which we got above that isn't the last result viewer or the last reserved So these are my significant independent variables that is product, product quality ecommerce advertising product lines exclusive image, competitive pricing packaging order billing price flexibility this I got in my step by selections for these set of variables. We have written the maximum value of adjusted R square. So, now I'm going to use the set of significant independent variables to predict the value of my dependent variable that is customer satisfaction for us. training data that is 70% of the data. I'm using the keyword output out of world to train period because the predicted values will be coming in the data set called train, train train which is going to be created inside work equal to predicted as a keyword predict the value of the customer satisfaction that is predict the value of the dependent variable and then run and then quit.

So that explains a whole bunch more. I have executed the regression procedure on the training data, I am predicting the value of customer satisfaction that is my dependent variable. So I'm building the model in that way only first have to write the name of the dependent variable that is customer satisfaction equals two These are my significant independent variables which I have got many last result that is in the stepwise selection there is product quality ecommerce advertising product line sales force image competitive pricing, packaging order billing price flexibility, these are the sets of independent variables which are significant to predict the value of my dependent variable output output train retrain create is a duplicate data set that is getting created inside the library work as I did not specify any library name it will be created and said work equals rented is a key word to predict the value of customer satisfaction which is our dependent variable then run and then quit.

So, now let's run this code Okay. See this is the result viewer. These are the independent variables that is a significant independent variables we need these parameter estimates value when we have to apply the results of our training data to our validation data sets. So these parameter estimates will actually be the conditions for our for our linear regression equation, which we are going to use in our validation data when do to predict the value of customer satisfaction for validation data set. So, this parameter estimates table is famous useful, we have to copy the coefficients from parameter estimates table for each of the independent variables. So, this is me Our spinner just pseudo see the adjusted R squared value has improved after doing the step by selection that is it is it is now at point one 4% and percent before now, it does improve three 2% This is my analysis of variance table.

And this model is built on 70% of the data that is training data. So, I've used so, the regression procedure is executed on 143 observations that is there were 143 observations that is aggregated to the training data because it is working on 70 doesn't have the data and just because we have generated random numbers therefore, the division will ever be exact it will be approx and every time you run this code you will get different sets of results. Now, let's view the data set to get the predicted value of a customer satisfaction for training data. This is the observed value of customer satisfaction, these are the values that was already their main data set. So, this is the predicted value of customer satisfaction, that is the primary value of the dependent variable for the training data. So now have to find the correlation between the observed value and predicted value of customer satisfaction for the training data.

So for that, I will be using the procedure called prop core prop Core Data equal to data Sydney will be trained grid My two variables will be one is the observed variable that is customer satisfaction. That is observable of my dependent variable and it is accredited. So you see this data set, these are the predicted values of the customer satisfaction. So the name of this V which is predicted. So I'm writing predicted. So I'm finding the correlation between this column that is predicted value of customer satisfaction and observed customer satisfaction that is this column, first column in the last column.

So let's find the correlation. It's run the code we are finding the correlation between these two variables. See you The religion is 0.90 which is quite good. That means the relative value in the observer, we're quite close to each other that means our prediction is correct for our training data. So, prediction for training data is quite close to the observed value of customer satisfaction for training data, the relative value of customer satisfaction, the observer of customer satisfaction, they are very close because my correlation coefficient is high that is 0.90. Now, we will be doing this much only in this video.

So, for now, let me end this video over here. Goodbye, and thank you. We'll see you on the next video.