Recall Problem 2.1 from page 70 in which you fit a simple linear regression model to data for 213 countries with response lntpct (percentage of the population that are internet users) and predictor gdp (GDPper capita in US$ thousands). This model is a reasonable one, but it is possible to improve it by transforming either the response variable or the predictor variable (or both). Investigate the use of transformations for this application using the data in the INTERNET data file. In particular, investigate natural logarithm and square root transformations for both the response and predictor variables. Which transformations seem to provide the most useful model for understanding any association between lntpct and gdp? Write up your results in a short report (no more than two pages) that compares and contrasts the different models that you fit.