Great Deal! Get Instant $10 FREE in Account on First Order + 10% Cashback on Every Order Order Now

Economics 102: Analysis of Economic Data Spring 2021 Department of Economics, U.C.-Davis Professor Diana Moreira Stata Take-Home Exam Version B Description: This is an open book exam. You may use any...

1 answer below »
Economics 102: Analysis of Economic Data Spring 2021
Department of Economics, U.C.-Davis
Professor Diana Moreira
Stata Take-Home Exam
Version B
Description:
This is an open book exam. You may use any class materials to help you. You can also use the
internet to get help with Stata commands. However, you must work independently. You are required to
submit your own work, and are not allowed to work together with others or get help from other people.
Please submit your exam as a single PDF. Clearly answer the questions below, inserting any relevant
Stata output directly into your answers. Additionally, you must turn in your .do file and .log file. Please
include these at the end of the same PDF with the .do file first and then the .log file. You will receive
2 points if you comply with this format.
For this exam, I have provided cross-sectional data on many variables that you can use. You must
use Data B (see
ief description below) Your answers will depend on the data that you use and
the variables you select for the answers below. You will all have unique answers since you will all choose
independently which variable you use. Some questions do not have a single co
ect answer, but you will
e graded on the quality of your answers and explanations.
Brief description of the dataset: Data B. Contains firm-level information about performance,
inputs and information about a firm’s staff. Each observation represents a different firm. Note: if there
are variables in log and you prefer to work in levels, just transform the variable using gen new=exp(old).
1. Use the command ”describe” and present the Stata output you obtained. How many variables are
there? and how many observations?
2. Propose a question about the association between two variables using this data (For example, from
our house price example, we asked how house price is associated with size or number of bedrooms).
(a) What is the question?
(b) Why is this question interesting?
(c) Describe how each of these two variables are measured in your data (you may transform
variables into new variables as needed, e.g., logs, rates or shares, etc.), what the unit of
observation of the data set is, and why this is cross-section data.
3. Using these two variables, specify which will be the dependent (y) variable and which will be the
independent (x) variables. Why?
4. Provide summary statistics using Stata for each of these variables. Comment. What summary
statistic do you think is most useful to know for each variable? Why?
5. Provide a histogram of your dependent variable.
(a) Choose either the number of bins or the bin width. Explain why you made this choice.
(b) Does the data appear skewed? Explain.
1
(c) Are there outliers? Please explain how can you identify the outliers and how many are there?
6. Run a bivariate regression using your dependent variable and the independent variable. Show the
egression output.
(a) Interpret the coefficients.
(b) Are the coefficients statistically significant? Explain.
(c) Predict the conditional mean of your dependent variable in Stata at the mean of your inde-
pendent variable
(d) Make a graph that includes (1) the actual data points (you can use a scatter plot of the
dependent and independent variable), (2) the prediction of the conditional mean (fitted values)
( Remember ‖ can be used to combine graphs with the same y and x variables in Stata.) Make
your graph look professional (something you would be proud to create for a client or boss) by
adjusting the formatting and including a title, axis labels, and any other desired components.
(Type “help graph intro” in Stata for an intro to making graphs with links to help with
making titles, labels, changing colors, etc. Or do a Google search.)
(e) What is the co
elation coefficient between your dependent and independent variable? (You
can use the “co
elate” command.) Explain how you could calculate the co
elation coefficient
using only the information from your bivariate regression table output.
(f) Interpret the co
elation coefficient
(g) Test the hypothesis that the true value of the slope coefficient is 20 percent higher than the
coefficient you estimated. Explain how you a
ived at your conclusion and use significance
level of 1%. [Hint: you will need to calculate the value for β∗ based on the value of you
estimated coefficient.]
7. Now I want you to continue using the same dependent variable and propose a new independent
variable. The new independent variable must be a variable that you believe is associated with the
dependent variable
(a) Explain what are the 3 variables you considered to be the ”new independent variable” and
the reason you chose the one you chose.
(b) Run a bivariate regression using your dependent variable and the new independent variable.
Show the regression output and interpret the slope coefficient.
(c) calculate the residual and present the summary command for the residual
(d) If you had to compare the model in item 6 and the model in 7.b, which one would you say
is better? i) Explain the criteria you used; ii) The interpretation of the criteria; iii) and a
limitation of this criteria.
(e) Suppose you could observe any variable that you want (not restricted to the variables in the
dataset). Can you propose a variable that you believe is associated with your dependent
variable and also with your independent variable (in model of item 7.b)?
(f) What does the existence of this variable imply for the slope coefficient estimated in 7.b? [hint:
think about the 4 assumptions for statistical inference and whether they will all hold]
Reminder: Please submit your exam as a single PDF. Insert any relevant Stata output directly
into your answers. Additionally, attach your .do file and .log file at the end of the same PDF with
the .do file first followed by the .log file.
2
Answered 1 days After May 16, 2021

Solution

Suraj answered on May 17 2021
138 Votes
Economics 102: Analysis of Economic Data Spring 2021
Department of Economics, U.C.-Davis
Professor Diana Moreira
Stata Take-Home Exam
This assignment is completed using the R studio. The solutions are given as follows:
Solution 1:
To check the dimensionality of the data set. We run the following function in R studio:
df<-read.dta("C:/Users/Hp/Desktop/data.dta")
dim(df)
The output is given as follows:
There are 100 observations and 37 different variables in the data set.
Solution 2:
a) We want to check that is there any association between log firm age and log sales variables.
) From the interest point of view, we want to check that whether with the increase in the long firm age the values of the log sales also increase or not.
c) Both the variables are measured in log values. This is cross sectional data because the data set observed by many variables of a firm at a single time.
Solution 3:
The dependent variable is that which depends on other variable and independent variable does not have any effect of other variable.
Here, in our question log firm age is the independent variable and log sales is the dependent variable. Because log sales values depend on log firm age.
Solution 4:
The summary statistics for both the independent and dependent variables is calculated as follows:
summary((log_firm_age))
Min. 1st Qu. Median Mean 3rd Qu. Max.
1.390 3.115 3.960 3.809 4.520 5.890
summary(log_sales)
Min. 1st Qu. Median Mean 3rd Qu. Max.
9.40 10.82 12.09 12.17 13.31 15.96
To describe about the best measure of summary statistic for each variable. The best way to check the distribution of the variable. Thus, histogram is the best way to determine about the distribution. The histogram for variables is given as follows:
Thus, we can see that the distribution of log firm age is skewed and for log sales it is approximately normal.
Thus, for skewed log firm age variable median is the best measure of central tendency and quartile deviation is best measure of deviation.
For log sales, mean is the best measure for central tendency and standard deviation is best measure for variability.
Solution 5:
a) The...
SOLUTION.PDF

Answer To This Question Is Available To Download

Related Questions & Answers

More Questions »

Submit New Assignment

Copy and Paste Your Assignment Here