Use Only For Class Purpose : For Any Question Please Use This Email: Andualem G, (PHD) Email

Download as pdf or txt
Download as pdf or txt
You are on page 1of 8

*****Use only for class purpose *****

Data Analysis has three parts

A. Descriptive analysis: It could have five parts

1. Description of variables: use tab and sum depending on the data


2. Relationship between the dependent and independent variables
I. Cross tabulation: It tries to see the existence of significant relationship
between dependent and independent variables. To check the existence of
significant relationship, look at the P value if it is below 10%.
II. Mean test: it tries to see the existence of mean difference in the continuous
variable (possibly the dependent variable) among the two groups(a dummy
variable). Check the left hand and the right hand t tests to decide on the
existence of significant mean difference in the continuous variable among the
two categories (the dummy variable). If the left hand side P value is below 10%
you can decide that there exists a significant mean difference among the two
categories.
III. Correlation: it uses to see the strength or weakness relationship among two
continuous variables. A value approach to one or negative one shows a strong
correlation among the two variables.
IV. Graph: it is optional, you can also show graphs to demonstrate or to describe
variables relationship
B. Econometrics Analysis: present the regression table and interpret the result
C. Qualitative analysis: For FGD and Interview part,

Please, check the following example

For any question please use this email: Andualem G, (PhD) Email: [email protected] Page 1
Determinants of Profit for Beer factories operating in Ethiopia
Chapter Four

4. data analysis and discussion


This paper is aimed to investigate the determinant factors of profit of beer factories operating in
Ethiopia. The paper studied all beer factories operating in Ethiopia. The paper used both descriptive and
econometrics analysis to achieve the objectives of the study.

4.1. Descriptive Analysis


The descriptive analysis used to describe and to check the significant relationship between profit and
independent variables.

4.1.1. Description of variables


The following table discussed categorical and continuous variables used in the paper.

Table 4.1: description of variables

Variable Variable Category Number Percent Mean Sd Max Min

Categorical variables Prize Yes - - - -

No - - - -

Location Amhara - - - -

Oromia - - - -

SNNP - - - ---

Tigray - - - -

ISO G-A

G-B

G-C

Technology High

For any question please use this email: Andualem G, (PhD) Email: [email protected] Page 2
Medium

low

Continuous variables Profit - -

Adv cost - -

Cost - -

Price - -

Shc

Source: own computation

******** please discuss each variable by considering its implication ******

4.1.2. The relationship between profit and independent


variables
In order to see the relationship between the dependent variable profit and independent variables, the
study used, cross tabulation, mean test and correlation. The following table shows the number of firms
which get positive profit and negative profit.

. tab pr

pr Freq. Percent Cum.

0 4 16.67 16.67
1 20 83.33 100.00

Total 24 100.00

the above table shows that there are 20 firms which get a positive profit and 4 of them incur a loss.

a. Cross tabulation between profit and Prize

For any question please use this email: Andualem G, (PhD) Email: [email protected] Page 3
. tab pr prize, chi2

prize
pr 0 1 Total

0 4 0 4
1 7 13 20

Total 11 13 24

Pearson chi2(1) = 5.6727 Pr = 0.017

Out of 20 firms which get a positive profit, 13 of them give prize for their customers and all firms who
incur a loss do not give prize. The probability value shows that, at 5% of significant level, we can reject
the null hypothesis that says there is no relationship between profit and prize. Hence, the study
concluded that, there is a significant relationship between profit and prize. Giving prize and getting a
positive profit are significantly related. .

b. Mean difference in profit for companies which provide


prize and which do not
To see, if there is a significant mean difference in the profit among the companies which give prize
and which do not, the study did mean difference test

Table 4.4: mean test for profit among …

Two-sample t test with equal variances

Group Obs Mean Std. Err. Std. Dev. [95% Conf. Interval]

0 11 -1700 1146.933 3803.945 -4255.525 855.5252


1 13 4422.077 757.6607 2731.785 2771.276 6072.878

combined 24 1616.125 910.4556 4460.303 -267.296 3499.546

diff -6122.077 1336.804 -8894.439 -3349.715

diff = mean(0) - mean(1) t = -4.5796


Ho: diff = 0 degrees of freedom = 22

Ha: diff < 0 Ha: diff != 0 Ha: diff > 0


Pr(T < t) = 0.0001 Pr(|T| > |t|) = 0.0001 Pr(T > t) = 0.9999

Source: own computation

For any question please use this email: Andualem G, (PhD) Email: [email protected] Page 4
The average profit of those firms which do not give prize is -1700 and those which give prize is 4422. The
difference in the mean income is -6122. The mean test statistics result shows that, for the left hand side
test, we can reject the null hypothesis (P value =0.0001) which says there is no mean difference in profit
among the two categories and concluded that the mean profit of firms which give prize is significantly
higher than the mean profit of firms which do not give prize. However, the right hand side test (P value =
99%) shows that we fail to reject the null hypothesis that says there is no difference in mean profit
among the two groups. Hence, we can conclude that, the mean profit of those who provide prize is
significantly higher than those which do not give prize.

c. Correlation between profit and cost


. pwcorr profit cost, sig

profit cost

profit 1.0000

cost 0.3507 1.0000


0.0929
The correlation between profit and cost is 0.35 and the significance level is 9.2%. Hence, at 10%
of significant level, the study concludes that there is a significant positive correlation between
profit and cost though the correlation is weak.

4.2. Econometrics Analysis


The aim of the study was to identify factors that affect the profit of firms. The study used Ordinary least
square model to identify factors that affect profit since profit is a continuous variable. The study used
price, advertising cost, prize, cost, location and ISO as independent variables. The regression table is
given in the following table:

Table 4.6: determinants of profit using OLS model

For any question please use this email: Andualem G, (PhD) Email: [email protected] Page 5
Source SS df MS Number of obs = 24
F( 9, 14) = 14.33
Model 412771406 9 45863489.5 Prob > F = 0.0000
Residual 44797658.7 14 3199832.77 R-squared = 0.9021
Adj R-squared = 0.8392
Total 457569065 23 19894307.2 Root MSE = 1788.8

profit Coef. Std. Err. t P>|t| [95% Conf. Interval]

price -77.87644 47.56787 -1.64 0.124 -179.8994 24.1465


advertising 286.488 122.4137 2.34 0.035 23.9368 549.0392
prize -6629.465 3455.575 -1.92 0.076 -14040.94 782.0069
cost -.003784 .0096873 -0.39 0.702 -.0245612 .0169931

location
2 -674.2274 2184.938 -0.31 0.762 -5360.454 4012
3 -1841.13 2796.908 -0.66 0.521 -7839.902 4157.641
4 -5929.819 2822.661 -2.10 0.054 -11983.83 124.1867

iso
2 5590.637 2194.565 2.55 0.023 883.7626 10297.51
3 5287.225 3322.559 1.59 0.134 -1838.955 12413.41

_cons 2678.007 3121.095 0.86 0.405 -4016.077 9372.09

Source: own computation

As we can see from the above regression table, all independent variables together are
significantly affects the dependent variable which is represented by Prob> F which is less than
1% and concluded that all independent variables together are determinant factors. The R2
measure is around 90% which is higher than the standard value of 50% and it implies that the
variation in the dependent variable profit can be explain by the variation in the independent
variables by 90%. Regarding to individual significant level, advertising cost, prize, location_4
and ISO_2 are significant. The above regression model can be written as follows:

Profit = 2678 -77 price +286 advertising-6629prize-0.0037cost ………

Before the discussion part, post estimation test is presented to be sure that the selected model is
consistent and efficient. The study undertakes post estimation tests of multicoliniarity test,
hetroscedesticity test, omitted variable test and normality test.
For any question please use this email: Andualem G, (PhD) Email: [email protected] Page 6
. ovtest

Ramsey RESET test using powers of the fitted values of profit


Ho: model has no omitted variables
F(3, 11) = 4.98
Prob > F = 0.0202

The omitted variable test checks the omission of important variables and the inclusion of
irrelevant variables. The probability value (0.42) indicates that, we fail to reject the null
hypothesis which says there is no omitted variable. Hence, we don’t have the problem of omitted
variable.
. hettest

Breusch-Pagan / Cook-Weisberg test for heteroskedasticity


Ho: Constant variance
Variables: fitted values of profit

chi2(1) = 18.71
Prob > chi2 = 0.0000

The probability value (P=0.300) indicates that we fail to reject the null hypothesis. Hence, we
don’t have the problem of heteroskedasticity..
. vif

Variable VIF 1/VIF

price 1.84 0.542275


advertising 13.06 0.076564
prize 22.24 0.044974
cost 4.57 0.218662
location
2 3.92 0.255340
3 12.12 0.082496
4 9.86 0.101461
iso
2 6.77 0.147645
3 18.40 0.054348

Mean VIF 10.31

The multicolinearity value is below 10%, hence we don’t have serial correlation problem.

For any question please use this email: Andualem G, (PhD) Email: [email protected] Page 7
. predict e, resid

. swilk e

Shapiro-Wilk W test for normal data

Variable Obs W V z Prob>z

e 24 0.86220 3.717 2.677 0.00371

The normality of error term is one of the guess markov assumptions for OLS estimation. As the
probability value is above 10%, we concluded that the error term is normal.

Coefficient interpretation and discussion


Advertising cost: is one of the determinant of profit and which is significant at 5%. The coefficient can
be interpreted as, when advertising cost increase by one birr profit increase by 286 Birr.
………………………………**** discussion should follow by comparing with other studies****………………..

Prize: is one of the determinant of profit and which is significant at 10%. The coefficient can be
interpreted as, using prize as means of promotion decrease profit by 6629 ………………………………****
discussion should follow by comparing with other studies****………………..

Location: is one of the determinant of profit and which is significant at 10%. The coefficient can be
interpreted as, firms in Tigray region got a profit of 5929 lower than firms located in Amhara region
………………………………**** discussion should follow by comparing with other studies****………………..

ISO: is one of the determinant of profit and which is significant at 5%. The coefficient can be interpreted
as, firms with ISO ranked B got a profit of 5590 higher than firms who are ranked with ISO A. This
indicates that ………………………………**** discussion should follow by comparing with other
studies****………………..

For any question please use this email: Andualem G, (PhD) Email: [email protected] Page 8

You might also like