Consider the output below from estimating a linear regression model using ordinary least squares (OLS) in Stata, which is based on a sample of 269 basketball players in the NBA (National Basketball Association) in the US. The outcome variable points is the number of points scored per game; forward is a dummy variable equal to 1 if a player's position on court is forward and zero otherwise; center is a dummy variable equal to 1 if a player's position on court is center and zero otherwise; guard is a dummy variable equal to 1 if a player's position on court is guard and zero otherwise; married is dummy variable equal to 1 if a player is married and zero otherwise; age measures a player's age in years; exper measures the total number of years spent as a professional basketball player; and expersq is the square of exper. Note that a player's position on court can only be forward, center, or guard. The XXX markers denote values in the Stata output which have been omitted, but which can be calculated based on the information provided. Answer all of the following questions using the Stata output above. 1. Explain why Stata has omitted the parameter estimate and standard error corresponding to the variable forward. [3 marks] 2. Interpret the parameter estimate on the variable age both in terms of its statistical significance and its magnitude. [3 marks] 3. Discuss whether there is any evidence that players in different positions score different numbers of points per game. [6 marks] 4. Holding fixed a player's position, marital status, and age, after how many years of experience does an additional year of experience actually reduce the number of points per game? [4 marks] 5. Propose a change to the regression if you suspect a differential effect of marital status on the number of points scored per game for black and non-black players. [4 marks]