Link to Data httpwwwfiledroppercomdataexcel2data Question 1

(Link to Data) http://www.filedropper.com/dataexcel2data

Question 1 I have uploaded an excel file named Data_Excel2.xls which has information on the following variables. State and congressional district identifier. democA records whether candidate A is a democrat or not. Similarly, V oteA gives the percentage of total votes that candidate A received. ExpendA and ExpendB record the total amount of campaign expenditure by each candidate in 1000s of dollars. What is the mean share of votes received by candidate A? What is the mean amount of expenditure by both candidates? What is the standard deviation of expenditures? - (5 Points)

Question 2 Using Excel, calculate the correlation between V oteA and ExpendA. What can you conclude about the relationship between these two variables? (5 Points)

Question 3 Now calculate the correlation between ExpendA and ExpendB. Is this sign of the relationship as you expected? (5 Points)

Question 4 Now run the following simple regression using Excel: V oteA = + ExpendA + u. 1. What are the estimated values of ˆ and ˆ? (5 points) 2. What is the interpretation of the coefficient on ExpendA? Is the sign of this relationship as you expected? (5 points) 3. What is the R2 ? Can you interpret this number? (5 Points) 4. Test the following hypothesis using the t-stat calculated by Excel. Ho : = 0; H1 : 6= 0 What do you conclude about ? (10 Points) 1

Question 5 Now run the following simple regression using Excel: V oteA = + ExpendB + u. 1. What are the estimated values of ˆ and ˆ? (5 Points) 2. What is the interpretation of the coefficient on ExpendB? Is the sign of this relationship as you expected?

Solution

Question 1

mean share of votes received by candidate A :   50.50

What is the mean amount of expenditure by both candidates? : 307.85

What is the standard deviation of expenditures

Variable StDev
expendA 281.0
expendB 306.3

correlation between V oteA and ExpendA

Pearson correlation of voteA and expendA = 0.395

they are not good linearly related