Scribd
Upload
45 views8 pages

Group Homework DH36DD01

1. The document contains a data table with sales volume (Y), price of product A (X2), and price of product B (X3). 2. Regression analyses were conducted to estimate the relationships between Y and X2, Y and X3, and Y with both X2 and X3. 3. The 3-variable regression model (Y = 40.13446 - 0.70559X2 + 0.600582X3) was found to best predict Y, based on it having a higher adjusted R-squared value than the 2-variable models and X3 being a statistically significant predictor of Y.

Uploaded by

Cát Đoàn Thảo Vy
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
45 views8 pages

Group Homework DH36DD01

1. The document contains a data table with sales volume (Y), price of product A (X2), and price of product B (X3). 2. Regression analyses were conducted to estimate the relationships between Y and X2, Y and X3, and Y with both X2 and X3. 3. The 3-variable regression model (Y = 40.13446 - 0.70559X2 + 0.600582X3) was found to best predict Y, based on it having a higher adjusted R-squared value than the 2-variable models and X3 being a statistically significant predictor of Y.

Uploaded by

Cát Đoàn Thảo Vy
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/ 8

GROUP

HOMEWORK
NUMBER NAME STUDENT ID

1 Trần Ngọc Ánh Phương 110302200021

2 Đặng Văn Khải 030136200857

3 Phạm Huỳnh Minh Tài 030136200538

4 Trần Ngọc Quỳnh 110302200022

5 Trần Đỗ Thúy Vy 110302200058


Y 46 48 42 39 29 24 28 28 21 30

X2 16 7 15 19 27 20 34 17 40 25

X3 17 14 23 13 13 11 22 14 9 12

There is a data table:


Y: Sales volume of a commodity A (Million product/month)
X2: Product A (USD)
X3: Price of product B (USD) related to product A

I. The population regression function Y relate to X2 has the following form:

Y =α 1 + α 2 X 2 + ε

1. Estimate the sample regression function.


Y = 50.22 - 0.76 X2
2. Whether variable Y depends on variable X2 or not with significance = 5%.
{ H 0 : α 1=0
H 1 :α 2 ≠ 0

When P value: 0.008 < 0.05


 Reject H 0
 There is no basis to reject H 0.
 Variable Y depends on variable X 2 .

II. The population regression function Y relate to X3 has the form:


Y =γ 1 +γ 2 X 3 + ν
. reg Y X3

Source SS df MS Number of obs = 10


F(1, 8) = 1.80
Model 148.662047 1 148.662047 Prob > F = 0.2163
Residual 659.837953 8 82.4797441 R-squared = 0.1839
Adj R-squared = 0.0819
Total 808.5 9 89.8333333 Root MSE = 9.0818

Y Coefficient Std. err. t P>|t| [95% conf. interval]

X3 .8901919 .6630667 1.34 0.216 -.6388427 2.419227


_cons 20.32516 10.225 1.99 0.082 -3.253724 43.90404

3. Estimate the sample regression function.


Y= 20.32516 + 0.8901919 X 3

4. Whether variable Y depends on variable X3 or not with significance = 5%.

{ H 0 : γ 1 =0
H 1 :γ 2 ≠ 0

When P-value: 0.082 > 0.05


 There is no basic to reject H 0.
 Variable Y does not depend on variable X3.

III. The population regression function Y relate to X2, X3 has the following
form:
Y = β1 + β 2 X 2 + β 3 X 3 +u
Source SS df MS Number of obs =
> 10
F(2, 7) =
> 7.71
Model 556.111167 2 278.055584 Prob > F = 0
> .0170
Residual 252.388833 7 36.0555476 R-squared = 0
> .6878
Adj R-squared = 0
> .5986
Total 808.5 9 89.8333333 Root MSE = 6
> .0046

>
Y Coefficient Std. err. t P>|t| [95% conf. inte
> rval]

>
X2 -.7055943 .2098961 -3.36 0.012 -1.20192 -.20
> 92688
X3 .6005823 .4467839 1.34 0.221 -.4558936 1.6
> 57058
_cons 40.13446 8.968179 4.48 0.003 18.92808 61.
> 34083

>
5. Estimate the sample regression function.
.
Y= 40.13446 - 0.70559 X 2 +0.600582 X 3

6. Test for statistical significance of regression coefficients of variables X2, X3


with significance level = 5%.

 { H 0 : β 2=β ¿2=0
H 1 : β 2 ≠ 0 , α=0.05

t (0.025 ,7) = 2.365

But: |t | > t(0.025 ,7)


3.36164 > 2.365
 Reject H 0.
 β 2 ≠ 0 means no statistical significance .
 { H 0 : β 3=β ¿3=0
H 1 : β 3 ≠0 ,α=0.05

t (0.025 ,7) = 2.365

But: |t | < t(0.025 ,7)


1.344235 < 2.365
 There is no basic to reject H 0.
 β 3=0 means statistical significance .

7. Test the fit of the model at the significance level = 5%.

{
2
H 0 : R =0
H 1 : R 2> 0

There is:
Prob (F) = 0 < 0.05
 Reject H 0.
 The model is suitable.

8. Estimate a 95% confidence interval for the coefficients and interpret its
significance.
α = 5% = 0.05

t (0.025 ,7) = 2.365

 ε^ t ^
(β ¿¿2 ¿)¿ ¿ = (0.025 ,7) * Se ( β 2) = 2.365* 0.209896 = 0.49640404

There is:
^
β 2 – ε^ β 2< ^
β 2 + ε^
(β ¿¿2 ¿)¿ ¿ < (β ¿¿2 ¿)¿ ¿

 - 0.70559 - 0.49640404 < β 2 < - 0.70559 + 0.49640404


 -1.20199 < β 2<−0.20918596
 β 2 ϵ (-1.20199; −0.20918596 ¿

ε^ ^
(β ¿¿3 ¿)¿ ¿ = (0.025 ,7) * Se ( β 3) = 2.365* 0.446784 = 1.05664416
 t

There is:
^
β 3 – ε^ β 3< ^
β 3 + ε^
(β ¿¿3 ¿)¿ ¿ < (β ¿¿3 ¿)¿ ¿

 0.600582 - 1.05664416 < β 3 < 0.600582 + 1.05664416


 - 0.45606186 < β 3 <1.65722616
 β 3 ϵ (0.45606186 ; 1.65722616 ¿

Conclusion:
- As X 2 increases by 1 unit, the mean of Y will decrease from 0.20918596 to
1.20199 units, holding the value of X 3 constant.
- As X 3 increases by 1 unit, the mean of Y will increase from - 0.45606186 to
1.65722616 units, holding the value of X 2 constant.

IV. To forecast for variable Y, should we choose a 2-variable or 3-variable


model? Give significance level = 5%.
Based on the results table
The 2-variable regression model (Y according to X2) has:
Y = 50.22 - 0.76 X2

Adjusted R-squared 0.5582

R3 variable > R2 variable


2 2

⇔ 0.5986 > 0.5582

{ 1
H : β =0
. H : β 0≠0 ,α
3
3
=0.05
P(X3)= 0.221 < 0.05
→ Reject H 0.
→ Y depends on X 3 .
→ Based on 3-variable model to predict Y.

V. Is it possible to know if the variable X2 or X3 affects Y more?


 It is possible to know if X 2 or X 3 affects Y more.
 It can be determined by the difference between the two values of X 2 and X 3
when put in the numbers of X 2 and X 3 to see the result.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy