ECON 6300/7320/8300: ELEMENTS OF ECONOMETRICS Tutorial 1: Stata and Basic Statistics
At the end of this tutorial you should be able to
• use Stata to read, manipulate and save data and workfiles
• use Stata to compute descriptive statistics
• use Stata to conduct hypothesis tests concerning a population mean
1 Introduction of Stata 1.1 Starting Stata
Before solving the problems, let’s look around the software. Stata can be started several ways. First, there may be shortcut on the desktop that you can double- click. For the Stata/SE Release 11 it will look like
Earlier versions of Stata have a similar looking Icon, but of course with a different number. Alternatively, using the Windows menu, click the Start → All Programs → Stata 11. A second way is to simply locate a Stata data file, with *.dta extension, and double-click.
1.2 Opening Display
Once Stata is started a display will appear that contains windows titled.
Command: Results: Review: Variables:
this is where Stata commands are typed
output from commands, and error messages, appear here a listing of commands recently executed
names of variables in data and labels (if created)
1
Across the top are Stata pull-down menus. We will explore the use of many of these. In the lower left-hand corner is the current path to a working directory where Stata saves graphs, data files, etc. We will change this in a moment.
1.3 Exiting
To end a Stata session click on File → Exit. Alternatively, simply type exit
in the Command window and press Enter. 1.4 Stata Data Files
Stata data files have the extension *.dta. These files should not be opened with any program but Stata. If you locate a *.dta file using double-click it will also start Stata. For the course, Econ 7310, data files and problem sets for each tutorial session will be available at the course webpage. For the exercise below we will use consumption.dta and fultonfish.dta. You should download the datasets into a convenient directory. To change the working directory use the pull-down menu File → Change Working Directory. In the resulting dialog box navigate to your preferred location and click OK. to this location type Stata will show the implied command
cd “C:\data\poe4stata” This can be entered into the Command window and press Enter.
2 Answer Key
Problems:
1. The text file consumption.dta contains observations on the weekly family consumption expenditure
(CONS) and income (INC) for a sample of 10 families.
(a) Read the data into Stata (Answer)
With Stata started, change your working directory to the where you have stored the Stata data files. In the Command window type
use consumption
and press Enter. If you have a data file already open, and have changed it in some way, Stata will reply with an error message.
no; data in memory would be lost
r(4)
If you click on r(4); you will be able to read the error message in a Viewer box. Sometimes this is helpful. To close the Viewer box click the X.
This feature will prevent you from losing changes to a data file you may wish to save. If this happens, you can either save the previous data file [more on this below], or enter the command
clear
The clear command will erase what is in Stata’s memory. If you want to open the data file and clear memory, enter
use consumption, clear
2
You can also open a Stata data file using the tool bar click the Open (use) icon on the Stata toolbar.
Locate the file you wish to open, select it, and click Open. In the Review window the implied Stata command is shown.
use “/Users/…/consumption.dta”, clear
In Stata opening a data file is achieved with the use command. The path of the data file is shown in quotes. The quotes are necessary if the path name has spaces included. The option clear indicates that any existing data is cleared from memory.
(b) Draw a scatter diagram of CONS against INC. (Answer)
From the pull-down menu, click on Graphics → two way graphs (scatter, line, etc)
Then, click on Create and choose Basic plots → Scatter
3
Then, under choose INC under X variable and choose CONS under Y variable; Then, click on Accept.
Then, you will see that “Plot 1” is ready under Plot definitions:.
By clicking on OK, you will see the scatter diagram;
4
In the Results window, you can find
twoway (scatter CONS INC)
is echoed, which means that by typing this expression in the Command window, you can generate the same diagram. Try it!
(c) On checking the data, you find that your assistant has recorded the weekly consumption expen- diture for Family 8 as $900 instead of $90. Correct this error and redraw the scatter diagram (Answer)
From the pull-down menu, click on Data → Data Editor → Data Editor (Edit)
Then, you will see a spreadsheet with data;
Go to the 8th row of the variable CONS, change the entry 900 to 90, and hit Enter. Finally, type twoway (scatter CONS INC)
in the Command window, which will generate the modified scatter diagram;
5
(d) Compute the mean, median, maximum and minimum values of INC and CONS. (Answer)
There are a few things you should do each time a data file is opened. First, enter the Command
describe
This produces a summary of the dataset in memory, including a listing of the variables, information about them, and their labels. A portion of the results is
Next, enter the Command
In the Results window we find the summary statistics. A portion is
where you find the mean, maximum, and minimum values of the variables. But, this result does not give you the median values. Alternatively, enter the Command
summarize CONS, detail
summarize
6
which gives a detailed distributional information on the variable CONS, including the median (50th percentile).
Activities:
Use the pull-down menus to obtain summary statistics. Also, explore the Commands
su
sum INC
summ INC, detail
summar, detail
What’s the difference between summarize and other commands here?
See if you can shorten the Command describe similarly, e.g., de?
Draw a histogram of CONS (Graphics → Histogram) and play around the main options (e.g., use 20 bins). See if you can draw the histogram using the Command window.
(e) Compute the correlation coefficient between CONS and INC. Comment on the result. (Answer)
Type in the Command window the following
correlate CONS INC
In the Results window we find the correlation matrix. A portion is
Activities:
Use the pull-down menus to have the same result. Statistics → Summaries, tables, tests → Summary and descriptive statistics → Correlations and covariances.
Discuss how to calculate covariance.
(f) Create the following new variables
DCONS = 0.5CONS
LCONS = log(CONS)
INC2 = INC2 √
SQRTINC = 7
INC
(Answer)
Type in the Command window
gen DCONS = 0.5 * CONS
generate LCONS = log(CONS)
gen INC2 = INC^2
gen SQRTINC = sqrt(INC)
(g) Delete the variable DCONS and SQRTINC from the workfile (Answer)
Type in the Command window
drop DCONS SQRTINC
(h) Delete this workfile. Type in the Command window (Answer)
exit, clear Activities: why do we need the option clear?
2. At the Famous Fulton Fish Market in New York city, sales of whiting (a type of fish) vary from day to day. Over a period of several months, daily quantities sold (in pounds) were observed. These data are in the file fultonfish.dat. Description of the data is in the file flutonfish.def. Describe the first four columns.
(a) Use Stata to open the data file and name the series in the first four columns as date, lprice, quan and lquan
(Answer)
fultonfish.dat is not in *.dta format. So, we can’t load it by typing
use fultonfish
When the data is not in Stata format, the way to import the file into the software depends on the structure/format of the data file. Fortunately, Stata offers a number of options to do this job. For this particular data file, from the pull-down menus, click on
File → Import → Text data (delimited, *.cvs, …)
8
Click on Browse and choose fultonfish.dat after setting File Format as All Files.
Then, set Delimiter: to Space and choose the option to treat sequential delimiters as one by checking the box next to Delimiter:. Then, in the preview you will see the data are nicely aligned with the default names such as v1, v2, v3, …. Notice that v1 is an empty column. So, the first variable appears in v2, the second in v3, and so on. Now, import the data by clicking on OK. Then, the Variables window shows something like;
Now, we change the variable names by typing in the Command window
rename v2 date
ren v3 lprice
ren v4 quan
ren v5 lquan
Whenever you change, check the Variables window. At the end, you must have
9
Since we will use only the four variables, type in the Command window keep date lprice quan lquan
(b) Compute the sample mean and standard deviation of the quantity sold (quan). (Answer)
In the Command window,
to have
su quan
So, the sample size is n = 111 and the sample mean X = 6, 334.667 and the sample standard deviation is σˆ = 4, 040.12.
(c) Test the null hypothesis that the mean quantity sold is equal to 7,200 pounds a day at the 5% level of significance.
(Answer)
The null is H0 : μ = 7,200 and the alternative is H1 : μ ̸= 7,200.
Since
Xn −μ0 6,334.67−7,200
σˆ/√n = 4, 040.12/√111 ≈ 2.26 > 1.96,
we reject H0
(d) Construct the 95% confidence interval for part (c)
(Answer)
6, 334.67 ± 1.96 × 4040.12/
√
111 = 6, 334.67 ± 751.58
(e) Label the variable lprice as “log(Price) of whiting per pound” and lquan as “log(Quantity)”. Then, plot lprice against lquan. Comment on the nature of the relationship between these two variables.
10
(Answer)
In the Command window, type
label variable lprice “log(Price) of whiting per pound”
label variable lquan “log(Quantity)”
twoway(scatter lprice lquan)
Then, the Variables window shows the labels for the variables;
Now, in the Command window, type
twoway(scatter lprice lquan)
and hit Enter. Then the diagram below will pop up;
Activities
Draw a histogram of lprice: Graphics → Histogram
(f) Save this workfile to any folder on any drive. (Answer)
File → Saveas
11