problem-10.11

problem-10.11  The predicted value can be found with predict(). Though you may not want to believe it, as the model has some issues. The simple plot of the residuals shows values that are scattered around 0, with nothing unusual. However, the second plot using the 1970 values on the x-axis instead of the index, shows that the variance increases with the price of the house.
> res = lm(y2000 ~ y1970, data=homedata)
> pred(res, newdata=dataframe(y1970=80000))
> predict(res, newdata=data.frame(y1970=80000))
[1] 321185
> plot(resid(res))              # simple plot
> plot(homedata$y1970,resid(res)) # shows spread