I am not using the answers as my own. I need help and guiadance with the answer.

2. The R data set at http://fsu.schmert.net/eco4421/data/MLB-attendance.RData has information about Major League baseball attendance for 32 teams over 1970-2000. Each of the 838 observations in the dataset represents one team in a single year -- for example,

i	city	nickname	season	home_attend	wins	runs_scored
1	Anaheim	Angels	1997	1767330	84	829
2	Anaheim	Angels	1998	2519107	85	787
3	Anaheim	Angels	1999	2253123	70	711
4	Anaheim	Angels	2000	2066977	82	864
5	Arizona	Diamondbacks	1998	3600412	65	665
...	...	...	...	...	...	...

(a) Make a scatterplot with runs_scored on the horizontal axis and home_attend on the vertical axis.

(b) Estimate a regression in which home_attend is the dependent variable and runs_scored is the independent variable. Report the results and interpret the coefficient values.

(c) Use the regression results to construct an 80% confidence interval for the effect of one more run per season on annual home attendance. [You can just do this calculation on a calculator. Use the results from the R printout and show how you calculated the interval.]

(d) How large does the sample estimate of _ have to be to reject

0:_=0in favor of:_>0 at a significance level of =.05. [Hint: this will depend on the standard error of the slope estimate, which you can see with the summary command.]

(e) Calculate the power of the test in part (d) for three cases

• _=100
• _=400
• _=1000

[Hint1: the power of the test is the probability that a sample will have a ^_ value positive enough to reject 0.] [Hint2: in R, you can calculate the probability that a z-variable is larger than X with the command 1-pnorm(X). For example, 1-pnorm(1) is Pr(Z>1), 1-pnorm(2.3) is Pr(Z>2.3), and so on.]

(f) Explain, in simple English, what the calculations in part (e) mean