Scribd
Upload
4 views3 pages

Practical 4

The document analyzes the relationship between test scores and sales made by trainees using simple linear regression. A significance test was conducted, resulting in the rejection of the null hypothesis, indicating the model is significant. The standard error of estimate was calculated to be approximately 10.966, confirming the model's good fit.

Uploaded by

Manjit Bhusal
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
4 views3 pages

Practical 4

The document analyzes the relationship between test scores and sales made by trainees using simple linear regression. A significance test was conducted, resulting in the rejection of the null hypothesis, indicating the model is significant. The standard error of estimate was calculated to be approximately 10.966, confirming the model's good fit.

Uploaded by

Manjit Bhusal
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/ 3

PRACTICAL:4

DATE:2081/11/04

Interested in determining the relationship between the test scores


and the sales made by the trainees at the end of one year in the
field. The following data were collected for 10 sales personnel who
have been in the field one year.
Sales
person 1 2 3 4 5 6 7 8 9 10
number
Test 2.6 3.7 2.4 4.5 2.6 5 2.8 3 4 3.4
scores(x)
Number
of units 95 140 85 180 100 195 115 136 175 150
sold(y)

a) Test of significance of goodness of fit of model at α=5%.


b) Compute standard error of estimate and interpret the result.

ANSWER:

a) Simple linear regression: Simple linear regression model is a


statistical model used to predict value of dependent variable using
independent value . It is given by
Y i= β^0 + β^1 X i +u^ i
Where,
Y =value of dependent variable
^
β 0 = estimated value of population Y-intercept β 0
^
β 1=estimated value of population slope β 1
X=value of independent variable

b) Goodness of fit: To test significance of goodness of fit of


regression model, we apply F distribution with df=(k,n-k-1)
where,
k=no of independent variables
n=sample size

H 0: β 1=0[Model is insignificant]
H 1 : β1≠ 0 [Model is significant]

Test statistics
Under H 0
MSSR
F= MSSE ~F(k,n-k-1)
Where,
k=no. of independent variable
Critical value:
For given α =5% , df=(k,n-k-1), we observe critical value of F-
statistic from F-table. i.e.
F tab

Decision
If F tab ≤ F cal , we reject H 0. Otherwise, we donot reject null
hypothesis.

b)
Standard error of estimate=√ MSSE
=
√ SSE
n−k −1

¿
√ Σ(Y i−Y^i)
n−k −1 √
= Σ u^ 2
n−k −1

Calculation:
a)
Model Summary
Std. Error
Mode R Adjusted R of the
l R Square Square Estimate
1 .963a .927 .918 10.961
a. Predictors: (Constant), X

ANOVAa
Sum of Mean
Model Squares df Square F Sig.
1 Regressio 12195.835 1 12195.835 101.51 .000b
n 9
Residual 961.065 8 120.133
Total 13156.900 9
a. Dependent Variable: Y
b. Predictors: (Constant), X

Here,
F calc > F tab
So, we reject null hypothesis i.e the model is insignificant.
Hence, the model is significant.

b)
Standard error of estimate=√ 120.133 =10.966

Conclusion
Hence , the model is a good fit.

You might also like