Scribd
Upload
12 views13 pages

Group

The document outlines a homework assignment for an Econometrics course, detailing various problems related to data analysis using EViews software. It includes instructions on creating workfiles, defining variables, running regression models, interpreting parameters, and calculating statistical measures such as SST, SSR, and SSE. Additionally, it discusses the importance of data types and provides steps for visualizing data through scatter plots and graphs.

Uploaded by

canqarazadeanar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
12 views13 pages

Group

The document outlines a homework assignment for an Econometrics course, detailing various problems related to data analysis using EViews software. It includes instructions on creating workfiles, defining variables, running regression models, interpreting parameters, and calculating statistical measures such as SST, SSR, and SSE. Additionally, it discusses the importance of data types and provides steps for visualizing data through scatter plots and graphs.

Uploaded by

canqarazadeanar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 13

MAHMUD AHMADZADA

KANAN KARIMOV
NIGAR HAJIZADA
KHAYALA ISLAMZADA
NAMIQ JAFAROV
ANAR JANGARAZADA
- GROUP 1000

Homework 2
00282 - Econometrics
Problem 1 (4 points)
Find variance of ^β 0.
Problem 2 (11 points)
1) Imagine you have collected observations on salary of CEO of 177
companies and their duration in the CEO position with their current
company in an excel file. Describe how you will create Eviews
workfile. (1 point)

1- Open EView on computer.


2- Go to File menu, select New and create a New Workfile.
3- We need some properties for processing of the workfile :
-Enter a name for our workfile in the Name
-We must choose the data type. Our example is cross-sectional type.
That is why, when creating workfile, we choose undated/unstructured
and enter 177 observations.
4- We need Import Data. For this, select File menu, and then click on
Import, after that Import From File and select the Excel file.
5- Define variables. For this, we must go to Quick section and from
here select the Define Variables menu. According to our example, our
variables are SALARY and CEOTEN.
6- Save. After creating the workfile, from File menu, we choose Save
and our workfile will be saved.

2) Please explain what is the data type. (0.5 points)

Data types are especially useful for the collecting, storing and analyzing
of economic data. They can come in a variety of types. The data type
defines how the data stored, interpreted, I mean which variables are
should considered importantly within the software. Types of the data
could be timely, ordinary, or just consist of some written variables. Data
type helps understanding how mathematical and statistical operations are
applied. Our example is cross-sectional data type.

3) Describe how you will create new series: logarithm of salaries


(Note: you already have salary in the data, you need to create its log
series) (1 point)
1-Open our workfile on EViews.
2-We select the existing salary variable. We already created the
workfile about 177 observations, so we do not need to create a new
one.
3-Go to series part. If there is not any series section, then from the
main menu we must select Quick or Proc. After this choose Make
Series.
4-In the series part we must write : series logsalaries.
This command will be equal to log(salaries).
All this happen with the help of Equation section.
5-In the final stage, click OK in order to create new seies. EViews will
show the new series with logarithm that we already entered before

4) Imagine you need to add new variable “knowledge of foreign


languages” – “forlang” of the CEO to the existing workfile, please
describe how you will do it? (1 point)

1-Open our workfile on EViews.We already created the workfile about


177 observations, so we do not need to create a new one.
2-Go to Object menu and select New Object.
4-In the Make Series we need to define the new series. After this, in the
Series Name field, we enter a name for our new series, such as
“forlang”.
5- In the Equation part, we define the value of “forlang” of the CEO.
6-In the final stage, click OK in order to create new seies. EViews will
show the new series with“forlang”that we already entered before

5) Please explain how you would draw the scatter plot where your
dependent variable is log(salary) and independent variable is
ceoten. (1 points)
In order to draw a scatter plot in EViews, we need to identify and
locate an independent variable ( ceoten ) and a dependent variable
( log(salary)).
1-Open our workfile on EViews with CEO slary data and the ceoten
variable.
2-We choose the LOG(SALARY) and CEOTEN variables at the same
time via holding down the CTRL key.
3-Creating the scatter plot. For this we must go to View menu, then
select the Graph menu (after selecting two variables and pressing the
right click of the mouse ) and from here select Scatter.
4-Defining the settings of the scatter plot.
Graph asks to us that we must identify the variables in the suitable X
and Y coordinate lines.
In the X line - CEOTEN; in the Y line - LOG(SALARY)
5-In the final stage, click OK and create the scatter plot.

6) Please describe how you would run the model: (1 point)

log ( salary )i =β 0+ β1 ceoteni +ui

1-Open your EViews workfile.Launch EViews and open the workfile


where you have your CEO salary data and the "Log_Salary" and
"ceoten" variables.
2-We must choose the dependent variable ( LOG(salary) ) and the
independent variable (CEOTEN ) by pressing the CTRL key.
3- Click the right click of the mouse and open as Equation. Choose the
method LS.
4-The equation format should look like this:
log(salary) = beta_0 + beta_1 * ceoten + u
5-Run the regression. After defining the equation and any additional
settings, click the "OK" button or the equivalent option to run the
regression analysis.

7) Results are as follows. Interpret parameters. Are the parameters


biased? Explain. If not, under which conditions they would be
biased? (1 points)
The parameters of a regression model are unbiased when the expected
value of the estimation errors is zero and when there is no correlation
between the independent variables and the errors.

8) Take the log of ceoten. Now run the model from Question 6 with
logceoten. Interpret results. Would you call it constant elasticity
model? Explain. (0.5 point)
9) Please explain how would you draw the graph of actual value, fitted
value and residuals (0.5 points).

1-Open our workfile on EViews.


2-We select the existing salary and ceoten variables. We already
created the workfile about 177 observations, so we do not need to create
a new one.
3-Open the selected variables as Graph (after choosing the View ) and
we choose the actual fitted residual graph.
4-Plot the actual values which are salaries on the Y line to observe the
real distribution of the data.
5-Plot the fitted values that are on X line compared to actual salaries.
6-For residuals (difference between actual and fitted values), we can
draw the second Y axis or draw another graph in order to show it. It
helps also identify the possible errors in our model.

10) Please explain how would you calculate manually, SST, SSR, SSE.
(1.5 points)

Total Sum of Squares (SST):


 SST measures the total variability in the dependent variable (Y).

 Yi is the observed value of the dependent variable for observation i.


 Yˉ is the mean of the dependent variable.

Regression Sum of Squares (SSR):


 SSR measures the variability in the dependent variable that is
explained by the regression model.
 Y^i is the predicted value of the dependent variable for observation
i obtained from the regression equation.
 Yˉ is, again, the mean of the dependent variable.

Error Sum of Squares (SSE):


 SSE measures the unexplained variability in the dependent
variable.

 Yi is the observed value of the dependent variable for


observation i.
 Y^i is the predicted value of the dependent variable for
observation i obtained from the regression equation.
These three components are related by the fundamental identity in
regression analysis:
SST=SSR+SSE
The goal in regression analysis is often to have SSR as large as possible
relative to SSE, indicating that the model is explaining a significant
portion of the variability in the dependent variable.

11) What is the value of σ – variance of the error? How would you
calculate/estimate it? (2 points)

The variance measures the spread between different numbers that located in the
data set. In other words, it shows how far each number locates from the mean
and other numbers in the data set.

The estimator SSR/n is biased essentially because it does not account for two
restrictions that must be satisfied by the OLS residuals. These restrictions are
given by the two OLS first order conditions:

One way to view these restrictions is this: if we know n - 2 of the residuals, we can
always get the other two residuals by using the restrictions implied by the first
order conditions. Thus, there are only n - 2 degrees of freedom in the OLS
residuals, as opposed to n degrees of freedom in the errors. It is important to
ˆ
understand that if we replace u i with ui in, the restrictions will no longer hold.

You might also like