Please submit your answers through the TurnitIn link on Blackboard. Put your name (or
student number) on your answers. If you worked in a group, please put down the names (or
student numbers) of other people in the group as well. Include relevant tables or graphs from your STATA output (you do not have to include the full log file or do file). Show your derivations. You will only get credit if you explain your answers.
1. (50 points) Download dataset ceosal17.dta from Blackboard and save it in your data folder
(or a folder where you keep STATA files). The dataset contains information about CEOs’
salaries, firms’ characteristics and their performance.
(a) Display the summary of the contents of the data set. How many variables and how many
observations are in the data set? What is the value of salary, roe, sales and finance in
the 10th observation (i.e. 10th row in the data set)?
(b) Display detailed summary statistics for the variable salary. What is the 25% percentile
(c) Calculate correlation between salary and return on equity (roe). What is the correlation
between salary and roe? Now generate a new variable salindol which records CEO’s
salaries in dollars, $, instead of thousands of $. What is the correlation between salindol
and roe? Compare it with the previous result. Is the value of the correlation coefficient
diffierent? Why? Explain.
(d) Compute the correlation matrix of variables salary, roe, and sales. What is the correla-
tion between sales and roe?
(e) Generate a variable nonperform which takes value 1 if the change in return on equity
(pcroe) is negative and 0 otherwise. What is the sample mean of nonperform? What is
the interpretation of this number?
(f) Display summary statistics of salary for each category of nonperform. Are the salaries
lower on average in non-performing firms? Are the salaries more variable in these firms?
(g) Create a new variable salary2 which is the square of the variable salary. Plot the variable
salary2 against salary.
(h) Display the histograms for return on equity and return on firm’s stock.
(i) Delete the variable salary2. Plot the variable salary against the variable sales. Put titles
on the axes and title the graph “CEO salaries and firm sales”.
(j) Generate a new variable hsalary which takes value 1 if CEO’s salary is above the median
(i.e. above 50% percentile) and 0 otherwise. Generate a new variable lowsalary which
takes value 1 if CEO’s salary is below or equal to the median and 0 otherwise. Display
two plots of the variable salary against the variable sales – one for CEO’s with high
salary and one for CEO’s with low salary. Make sure that you label the axes and put
titles on your graphs. Is the relationship between salaries and sales linear in both of
these groups? (i.e. can you approximately draw a straight line through the data?)
EasyDue™ 支持PayPal, AliPay, WechatPay, Taobao等各种付款方式!
E-mail: email@example.com 微信:easydue