Often times, the dependent variable is transformed using the natural logarithm function, ln(y) in order for the assumption of normality to be more reasonably satisfied (price data are oftentimes very non-normal). If we let Y* = ln(Y); then it can be shown that (eβ1–1)100 is now the estimated percentage increase in Y for a one-unit increase in the dependent variable, where e ≈ 2.718 is the base of the natural logarithm. Use this fact in your interpretations below.
(a) Create a new variable named "lnSale" by taking the natural logarithm of SalePrice000s. Note that the natural logarithm function in R is log. In Excel, the function is LN(). Make a scatter plot of lnSale by TotalSF and comment on the appearance of this scatter plot.
(b) Write down the first-order multiple linear regression model to predict Sales000s using "age" and "totalsf". Again use the actual variable names in place of Y and x: Then .t the regression line, using Excel, R, or any other program you want. Display the output from whatever program you use (a screen shot will probably be easiest), and write down the estimated regression equation.
(c) Interpret the estimated slope coefficients in the context of the problem.
The question belongs to Statistics and it discusses about producing a scatterplot and interpreting the data given in the Excel sheet.
Total Word Count 197Download Full Solution