TOPIC 3: DATA ANALYSIS IN EVIEWS

EViews is the short form for Econometric Views, and hence it is an application that is most

suited for econometric data analysis, especially the analysis of time series data. However, it can

also work with cross sectional and panel data as well.

EViews can be used for the following:

1) data analysis and evaluation

2) financial and cost analysis

3) macroeconomic forecasting

4) simulations

Just like Stata, EViews provides the user with flexibility since it has both a command window

and click-and-point interface menu.

The EViews application when opened shows five windows as follows:

Eviews

File Edit Object View Proc Quick Options Add-ins Window Help

Welcome to Eviews

The window containing: File, Edit, Object, View, Proc, Quick, Options, Add-ins, Window and

Help is known as the MENU BAR.

WORK AREA, and finally the STATUS BAR with the words “Welcome to EViews”.

EXAMPLE TO DEMONSTRATE HOW TO WORK WITH EVIEWS

STEP 1: TYPE IN THE FOLLOWING DATA (LONGLEY DATA 1947 1962) IN EXCEL

THEN SAVE ON YOUR DESKTOP

Table 1: The Longley Dataset 1947 to 1962

OBS Y X1 X2 X3 X4 X5 X6

1947 60323 830 234289 2356 1590 107608 1

1948 61122 885 259426 2325 1456 108632 2

1949 60171 882 258054 3682 1616 109773 3

1950 61187 895 284599 3351 1650 110929 4

1951 63221 962 328975 2099 3099 112075 5

1952 63639 981 346999 1932 3594 113270 6

1953 64989 990 365385 1870 3547 115094 7

1954 63761 1000 363112 3578 3350 116219 8

1955 66019 1012 397469 2904 3048 117388 9

1956 67857 1046 419180 2822 2857 118734 10

1957 68169 1084 442769 2936 2798 120445 11

1958 66513 1108 444546 4681 2637 121950 12

1959 68655 1126 482704 3813 2552 123366 13

1960 69564 1142 502601 3931 2514 125368 14

1961 69331 1157 518173 4806 2572 127852 15

Source: J Longley, “An appraisal of least squares programs from the point of view of the user,”

Journal of the American Statistical Association, vol. 62, 1967, pp 819 – 841

Domador N Gujarati and Dawn C Porter, Basic Econometrics, 5th edition, McGraw Hill

Edition 2009, p. 348

The Longley data given above are time series for the years 1947 – 1962 and pertain to: Y =

number of people employed, in thousands; X1 = GNP price deflator; X2 = GNP in millions of

dollars; X3 = number of people unemployed, in thousands; X4 = number of people in the armed

forces; X5 = noninstitutionalised population over 14 years of age; and X6 = year, equal to 1 in

1947, 2 in 1948 and so on up to 16 in 1962.

STEP 2: CREATING A WORKFILE

The first step in using EViews is to create a work file, that is, a container for all EViews Files.

To create a work file in EViews, go to the menu bar and select: file/new/work file

A dialogue box will pop up asking you to specify a few details about your work file. You may

select as follows:

- Work file structure: dated-regular frequency

- Frequency annual

- Start date 1947

- End date 1962

- WF Longley Data
Then click OK.

By doing so, you will already have created the work file. Notice that the work file has two icons:

an icon labeled “c” which stands for intercept, and another icon labeled “resid” which stands for

residuals.

There are three work file structure types:

- unstructured or undated – this is used for cross sectional data

- dated-regular frequency – this is used for time series data

- Balanced panel – this is used for balanced panel data.

STEP 3: SAVING THE WORKFILE

To save the work file created, click as follows:

Menu/file/save as/save, select double precisions then click OK

STEP 4: IMPORTING DATA

To import data in EViews, select: file/import/import from file

EViews will ask you to specify the location of your data, for example: DESKTOP – EVIEWS

DATA ANALYSIS – THE LONGLEY DATA 1947 1962.

Then click: Next, Next, Finish.

By doing so, you will be able to see that new icons are created on the work file as follows: x1,

x2, x3, x4, x5, x6 and y. Thus, you have already imported the data.
STEP 5: VERIFYING THE DATA

On your work file, highlight: x1, x2, x3, x4, x5, x6 and y and then click the left button of mouse

and select: open/as group

This process displays the spreadsheet for your data. The dataset consists of 7 variables and the

observations are from 1947 to 1962. Thus, compare the EViews spreadsheet with the original

spreadsheet.

STEP 6: GRAPHICS

To draw a line graph for the variables:

Menu/view/graph/basic type/line and symbol/multiple graphs, then click OK.

From the graphs, we notice that our variables have an upward trend. The upward trend is smooth

for some but it is volatile for other variables.

To draw a bar graph, proceed as follows:

Menu/view/graph/basic type/bar/multiple graphs, then click OK.

To get a scatter diagram:

Menu/view/graph/basic type/scatter/multiple graph, then click OK.

STEP 7: DESCRIPTIVE STATISTICS

other summary statistics.

To obtain the descriptive statistics: Menu/view/descriptive statistics/common sample

STEP 8: GENERATING NEW VARIABLES

To create the square of X1: Menu/Quick/Generate Series, then type as follows: squareX1 =

X1*X1, then click OK

To create the square-root of X2: Menu/Quick/Generate Series, then type as follows: sqrootX2 =

X2 ^ 0.5, then click OK

To create the logarithm of X3: Menu/Quick/Generate Series, then type as follows: logX3 =

log(X3), then click OK

To create the first lag of X4: Menu/Quick/Generate Series, then type as follows: dX4 = X4(-1),

then click OK

STEP 9: ESTIMATING A REGRESSION EQUATION

To estimate an ordinary least squares (OLS) regression model, then select as follows:

Quick/estimate Equation

A dialogue box appears which requires you to specify the following:

- Equation specification y c x1 x2 x3 x4 x5 x6

- Method LS – Least Squares

- Sample 1947 1962

The result is the regression model:

Dependent Variable: Y

Method: Least Squares

Date: 06/22/14 Time: 14:36

Sample: 1947 1962

Included observations: 16

Variable Coefficient Std. Error t-Statistic Prob.

C 77270.12 22506.71 3.433204 0.0075

X1 1.506187 8.491493 0.177376 0.8631

X2 -0.035819 0.033491 -1.069516 0.3127

X3 -2.020230 0.488400 -4.136427 0.0025

X4 -1.033227 0.214274 -4.821985 0.0009

X5 -0.051104 0.226073 -0.226051 0.8262

X6 1829.151 455.4785 4.015890 0.0030

R-squared 0.995479 Mean dependent var 65317.00

Adjusted R-squared 0.992465 S.D. dependent var 3511.968

S.E. of regression 304.8541 Akaike info criterion 14.57718

Sum squared resid 836424.1 Schwarz criterion 14.91519

Log likelihood -109.6174 Hannan-Quinn criter. 14.59449

Prob(F-statistic) 0.000000

To interpret the coefficients, they show how a unit change in any variable affects the dependent

variable, holding other variables constant.

For example, the coefficient on x1 is 1.506187. This means that a unit increase in x1 leads to an

increase in y by 1.5062 units, holding the other variables constant. The other coefficients are

interpreted in a similar way.

ACTIVITY:

Perform the following regression equation and interpret the results of your model

Regress Y c X1 X2 X3 X4 X5 X6 squareX1 sqrootX2 logX3 dX4

MULTICOLLINEARITY PROBLEM

The R squared is very high (0.9955) yet some variables are insignificant since the prob. Values

are greater than 0.1. The variables x1, x2 and x5 are therefore not significant. If a model has a

high R-Squared and most variables insignificant, then MULTICOLLINEARITY is a very likely

problem. Recall that multicollinearity is one of the violations of the OLS assumptions.

However, to formally test for this, we can use the following methods:

(i) The correlation matrix

To obtain the CORRELATION MATRIX in EViews, we proceed as follows: Quick/group

statistics/correlations. On the series list that pops up, click OK. You can also click FREEZE in

the Menu bar. The output below is the correlation matrix.

X1 X2 X3 X4 X5 X6 Y

X1 1.000000 0.991589 0.620633 0.464744 0.979163 0.991149 0.970899

X2 0.991589 1.000000 0.604261 0.446437 0.991090 0.995273 0.983552

X3 0.620633 0.604261 1.000000 -0.177421 0.686552 0.668257 0.502498

X4 0.464744 0.446437 -0.177421 1.000000 0.364416 0.417245 0.457307

X5 0.979163 0.991090 0.686552 0.364416 1.000000 0.993953 0.960391

X6 0.991149 0.995273 0.668257 0.417245 0.993953 1.000000 0.971329

Y 0.970899 0.983552 0.502498 0.457307 0.960391 0.971329 1.000000