The file P17_23.xlsx contains data on 80 (fictional) houses, including their selling prices, although the selling price is missing for the first house. It turns out that when the dependent variable is numeric, NeuralTools can compare a linear regression with a neural net. If the regression has a lower RMSE than the neural net on the test cases, then the regression equation is used for prediction; otherwise, the neural net is used for prediction. Run NeuralTools on this data set. In the Training dialog box, accept all the defaults, including the option to Perform Linear Regression on the Net Configuration tab, but check the Place Predicted Values Directly in Data Set option on the Train tab. You will get a prediction of the selling price of the first house on the Data sheet. Can you explain, in detail, how this prediction is calculated?