Amy asked me though why I needed this information. "I will tell you later-, I responded. .
There was no way for me to take the whole population of 90 houses out of John's list and examine all of those. I decided to take a random sample of 50. .
Aware that the prices displayed in the old data were not representative for the present time, I decided to double those (processed "=B5*2- and copied this command for all the other effected cells). Now I was left with this sample source.
.
Soon after, a discussion between the two of us started:.
Amy asked me gain, "What do you need this long list for?-.
I answered: "I will apply multiple linear regression to predict the prices we will need to place our house-. .
Just as every woman she did not want to stop asking: "What is multiple regression?- .
My answer was: " Multiple regression is used to measure the relationship between one dependent variable and several independent variables. In our case the dependent variables are the Asking Price, Selling Price and the Difference between the Asking and Selling Price (discount), which are the "y-values-. But we need to be careful, the independent variables (in our case the attributes of the houses) can predict the dependent variable, but the dependent variable cannot be used to predict the independent variables. It is a one-way analysis. In addition, independent variables selected should have strong correlations with the dependent variable but only weak correlations with other independent variables-.
.
However, I questioned whether the "x-values- in our case were really the ones, which were determinants of house prices. My wife insisted on keeping those amenities she had chosen for my analysis and I had to obey. But still, I was not sure whether the independent variables we were using were justified theoretically.
Amy: -When do people use multiple regression?-.
Ozan: "Multiple regression is used when the researcher wants to determine what variables contribute to the explanation of the dependent variable and to what degree.
