Tugas 1 Ekman

6.
this is the result that you obtain from regression.
(a) the estimated regression line is

Y = 114.073 -9.4701*X1 + 0.0289*X2
(b) test at 5% for significance of slope parameters. we will do one side interval test
because demand is negatively related with price and income is positively related to
price (assuming normal goods).
from the above table we see the |t| > 4
we can directly apply rue of thumb and say that our results are statistcally
significant.
rule of thumb here is that if the mod of calculated value exceeds 3, we reject our null
hypothesis.
(c) as observed from the table.

the coefficient of determination is 0.968
adjusted coefficient of detremination is 0.964
(d) we check f statistic for overall significance of regression.

our test statistic here is 258 which is very high so we can reject our null hypothesis
without even finding critical value.
6. From the following table giving the quantity demanded of a commodity (Y), its
price (X1), and consumers' income (X2) from 1981 to 2000, (a) estimate the
regression equation of Yon X, and X2. (b) test at the 5 percent level for the statistical
significance of the slope parameters, (c) find the unadjusted and the adjusted
coefficients of determination, and (d) test at the 5 percent level' for the overall.
statistical significance of the regression. Show all your results to three decimal
places. [If you have not used the computer before, ask á fellow student to show you
how to run the regression. If this is not possible, from the answer to part (a) of this
question provided at the end of the book, answer parts (b), (c), and (d).)
Answer:
Enter the data into Excel.
Go to Data > Data Analysis > Regression.

Input the data range and check Labels.
Click OK. The output is as follows:
(a) The regression equation is:
Y = 114.074 - 9.470X1 + 0.029X2
(b) The hypothesis being tested is:

H0: β1 = 0
H1: β1 ≠ 0
The test statistic, t = -5.2046
The p-value is 0.000.
Since the p-value (0.000) is less than the significance level (0.05), we can reject the
null hypothesis.
Therefore, we can conclude that the slope is significant.
The hypothesis being tested is:
H0: β2 = 0
H1: β2 ≠ 0
The test statistic, t = 4.506
null hypothesis.
Therefore, we can conclude that the slope is significant.
(c) Unadjusted coefficient of determination = 0.968
Adjusted coefficient of determination = 0.964
(d) The hypothesis being tested is:
H0: β1 = β2 = 0
H1: At least one βi ≠ 0
The test statistic, F = 258.942
null hypothesis.
Therefore, we can conclude that the model is significant.

6. Regression analysis in excel:
In excel- go to Data - Data analysis tool pack- click on Regression- Input y range and
x range & confidence level- give output range then click ok.
We get following output:
ANOVA table:
a) Regression equation:
commodity= 113.105-9.383 (Price) + 0.029 (Consumer income)

b) At %5 level of significanece, regression coefficient of X1 is statistical significant
because p value is 0.000 and it is less than level of significance.
P value of Regression coefficient of X2 is 0.000 which is less than level of

significane, hence regression coefficent is statisticaly significant.
c) From the output of regression:
Coefficient determination= 0.967
Adjusted coefficient of determination = 0.964
d) From ANOVA table, F- test statistic is 252.940 and F critical is 2.251E-13, here F
critical is less than F test statistic hence we conclude that overall regression
coefficient are statisticaly significant.
7. If the regression of Y on X1 and X2 is run in double-log form the data of problem

6, the result are as follows: ln Yt=-0,533-0,389 ln X1t+0,769 ln X 2t. Rata-rata R^2
= 0,95054 F=183,582. Compare the above result with those of problem 6. Which are
better? why?
Answer:
Comparing the results from problem 6 with problem 7:
Problem 6:
 The estimated regression line is Y = 114.073 - 9.4701X1 + 0.0289X2

 Coefficient of determination (R-squared) is 0.968
 Adjusted coefficient of determination is 0.964
 F-statistic for overall significance of regression is 258
Problem 7:
 The estimated regression line in double-log form is ln Yt = -0.533 - 0.389 ln

X1t + 0.769 ln X2t
 Coefficient of determination (R-squared) is 0.95054
 F-statistic for overall significance of regression is 183.582
Comparing the R-squared values of the two problems, we see that the value in
problem 6 is higher than problem 7 (0.968 > 0.95054). This suggests that the
regression in problem 6 explains a greater proportion of the variation in the data than
the regression in problem 7.
However, when it comes to overall significance of regression, the F-statistic in
problem 6 is much higher than that in problem 7 (258 > 183.582). This means that
the regression in problem 6 is more significant overall.
As for which results are better, it depends on the specific context and goals of the
analysis. If the main goal is to explain a large proportion of the variation in the data,
then the results from problem 6 may be more desirable. On the other hand, if the
main goal is to test for overall significance of the regression, then the results from
problem 7 may be more relevant.
Membandingkan hasil dari masalah 6 dengan masalah 7:
Masalah 6:
 Garis regresi yang diestimasi adalah Y = 114,073 - 9,4701X1 + 0,0289X2

 Koefisien determinasi (R-kuadrat) adalah 0,968, artinya 96,8% mampu
menjelaskan variasi dari variabel dependen.
 Koefisien determinasi disesuaikan adalah 0,964
 Statistik F untuk signifikansi regresi secara keseluruhan adalah 258
Masalah 7:
 Garis regresi yang diestimasi dalam bentuk log-log ganda adalah ln Yt = -

0,533 - 0,389 ln X1t + 0,769 ln X2t
 Koefisien determinasi (R-kuadrat) adalah 0,95054, artinya 95% model
mampu menjelaskan variasi dari variabel dependen.
 Statistik F untuk signifikansi regresi secara keseluruhan adalah 183,582
Membandingkan nilai R-kuadrat dari kedua masalah, kita melihat bahwa nilai pada
masalah 6 lebih tinggi dari masalah 7 (0,968 > 0,95054). Ini menunjukkan bahwa
regresi pada masalah 6 menjelaskan proporsi variasi dalam data yang lebih besar
daripada regresi pada masalah 7.
Namun, ketika datang ke signifikansi regresi secara keseluruhan, statistik F pada

masalah 6 jauh lebih tinggi daripada pada masalah 7 (258> 183,582). Ini berarti
bahwa regresi pada masalah 6 lebih signifikan secara keseluruhan.
Adapun hasil yang lebih baik tergantung pada konteks dan tujuan analisis yang
spesifik. Jika tujuan utama adalah menjelaskan proporsi variasi dalam data yang
besar, maka hasil dari masalah 6 mungkin lebih diinginkan. Di sisi lain, jika tujuan
utama adalah menguji signifikansi regresi secara keseluruhan, maka hasil dari
masalah 7 mungkin lebih relevan.

Tugas 1 Ekman

Uploaded by

Tugas 1 Ekman

Uploaded by

6.

this is the result that you obtain from regression.

(a) the estimated regression line is

(c) as observed from the table.

(d) we check f statistic for overall significance of regression.

Enter the data into Excel.

Go to Data > Data Analysis > Regression.

Click OK. The output is as follows:

(a) The regression equation is:

Y = 114.074 - 9.470X1 + 0.029X2

(b) The hypothesis being tested is:

The test statistic, t = -5.2046

The p-value is 0.000.

Therefore, we can conclude that the slope is significant.

The hypothesis being tested is:

The test statistic, t = 4.506

The p-value is 0.000.

Therefore, we can conclude that the slope is significant.

(c) Unadjusted coefficient of determination = 0.968

Adjusted coefficient of determination = 0.964

(d) The hypothesis being tested is:

H0: β1 = β2 = 0

H1: At least one βi ≠ 0

The test statistic, F = 258.942

The p-value is 0.000.

Therefore, we can conclude that the model is significant.

We get following output:

commodity= 113.105-9.383 (Price) + 0.029 (Consumer income)

P value of Regression coefficient of X2 is 0.000 which is less than level of

c) From the output of regression:

Coefficient determination= 0.967

Adjusted coefficient of determination = 0.964

7. If the regression of Y on X1 and X2 is run in double-log form the data of problem

Comparing the results from problem 6 with problem 7:

 The estimated regression line is Y = 114.073 - 9.4701X1 + 0.0289X2

 The estimated regression line in double-log form is ln Yt = -0.533 - 0.389 ln

Membandingkan hasil dari masalah 6 dengan masalah 7:

 Garis regresi yang diestimasi adalah Y = 114,073 - 9,4701X1 + 0,0289X2

 Garis regresi yang diestimasi dalam bentuk log-log ganda adalah ln Yt = -

Namun, ketika datang ke signifikansi regresi secara keseluruhan, statistik F pada

You might also like