The following regression is fitted using variables identified that could be rela
ID: 3182328 • Letter: T
Question
The following regression is fitted using variables identified that could be related to student requested loan amount , LOAN-AMT($) for returning students of a certain University.
LOAN-AMT = + ACCEPT + PREV + OUTS
Where ACCEPT = the percentage of applicants that was accepted by the university, PREV = previous loan amount and OUTS =outstanding loan amount
The data was processed using MNITAB and the following is an extract of the output obtained:
The regression equation is LOAN-AMT = -26780 + 116.00ACCEPT – 4.21PREV + 70.85OUTS
Predictor Coef StDev T P
Constant - 26780 6115 * 0.000
ACCEPT 116.00 37.17 3.14 0.003
PREV - 4.21 14.12 -0.30 **
OUTS 70.85 15.77 4.49 0.000
S = 2685
R-Sq = 69.6%
R-Sq(adj) = 67.7%
Analysis of Variance
Source DF SS MS F P
Regression 3 808139371 269379790 37.37 0.000
Residual Error 49 353193051 7208021
Total 52 1161332421
a) What is the sample size used in this investigation?
(b) Fill in the blanks identified by ‘*’ and ‘**’.
(c) Is significant, at the 5% level of significance?
(d) Is significantly different from 120?
(e) If a F-Test is to be conducted, state the appropriate null and alternative hypotheses for this test.
Explanation / Answer
a) DF(Total) = 52, where the equation is n - 1 where n is the sample size
n = 53
b) Tconstant = -26780/6115 = -4.38
Pt=-0.30,49 = 0.765
c) p-value = 0.003 < 0.05, therfore it is significant
d) t = (-4.21 - 120)/14.12 = -8.80
p-value = 0.000. It is significantly different from 120
e) H0: = = = 0
H1: Any of , or 0