Bayesian Logistic Regression Models for Credit Scoring by Gregg Webster


Bayesian logistic regression model with an informative prior



Download 2,26 Mb.
Pdf ko'rish
bet42/58
Sana08.07.2022
Hajmi2,26 Mb.
#757017
1   ...   38   39   40   41   42   43   44   45   ...   58
Bayesian logistic regression model with an informative prior
 
Prior information came from the model fitted on the “old” data. The model fitted on the 
“old” data serves as expert information obtained in the home country. This expert 
knowledge on the logistic regression parameters was then used as prior information for the 
model on the limited amount of “new” data in the new economic location. A multivariate 
normal prior is assumed for the parameters. The prior parameters are also assumed to be 
independent. The prior coefficients are the coefficients from the logistic regression on the 
“old” data. Each coefficient has corresponding information represented in a 17 x 17 
diagonal matrix. The prior coefficients and corresponding element in the diagonal matrix 
are given in Table 4.12
Table 4.12
Prior parameters for an informative Bayesian logistic regression model. 
 Variable 
Coefficient 
Information 
(Intercept) 
-7.194241 
2.88E+02 
LOAN 
-2.3673E-05 
1.19E+11 
MORTDUE 
-3.70998E-06 
1.94E+12 
VALUE 
3.03441E-06 
3.80E+12 
REASONHomeImp 0.2027903 
9.37E+01 
JOBOffice 
-0.681924 
3.90E+01 
JOBOther 
0.01722975 
1.36E+02 
JOBProfExe 
0.04760004 
5.46E+01 
JOBSales 
0.4024014 
6.64E+00 
JOBSelf 
0.4016077 
8.87E+00 
YOJ 
-0.01615048 
3.20E+04 
DEROG 
0.7334939 
1.90E+02 
DELINQ 
0.8039918 
3.56E+02 
CLAGE 
-0.005222989 8.85E+06 
NINQ 
0.1366665 
1.70E+03 
CLNO 
-0.02814893 
1.54E+05 
DEBTINC 
0.1911389 
4.13E+05 


86 
In order to get the acceptance rate between 20-40%, a tuning parameter of 0.6 was used. 
Because a high dimension model is being fitted, the acceptance rate needs to be towards 
the lower bound of the desired range. The tuning parameter of 0.6 gave an accepted rate of 
23%. The model is summarized in Table 4.13. 
Table 4.13
Bayesian logistic regression model on the “new” data with an informative 
prior. 
The mean provides the estimate for the parameter. From Table 4.13, looking at the 
quantiles for each variable we can determine which variables are significant at the 5% 
significance level. The values from the 2.5% to the 97.5% quantiles provide a 95% 
credibility interval for each variable. Only dummy variables for the JOB variable and 
REASON variable are insignificant. This shows that the majority of variables included in 
the model are significant in predicting good and bad applicants. The parameter estimates 

Download 2,26 Mb.

Do'stlaringiz bilan baham:
1   ...   38   39   40   41   42   43   44   45   ...   58




Ma'lumotlar bazasi mualliflik huquqi bilan himoyalangan ©hozir.org 2024
ma'muriyatiga murojaat qiling

kiriting | ro'yxatdan o'tish
    Bosh sahifa
юртда тантана
Боғда битган
Бугун юртда
Эшитганлар жилманглар
Эшитмадим деманглар
битган бодомлар
Yangiariq tumani
qitish marakazi
Raqamli texnologiyalar
ilishida muhokamadan
tasdiqqa tavsiya
tavsiya etilgan
iqtisodiyot kafedrasi
steiermarkischen landesregierung
asarlaringizni yuboring
o'zingizning asarlaringizni
Iltimos faqat
faqat o'zingizning
steierm rkischen
landesregierung fachabteilung
rkischen landesregierung
hamshira loyihasi
loyihasi mavsum
faolyatining oqibatlari
asosiy adabiyotlar
fakulteti ahborot
ahborot havfsizligi
havfsizligi kafedrasi
fanidan bo’yicha
fakulteti iqtisodiyot
boshqaruv fakulteti
chiqarishda boshqaruv
ishlab chiqarishda
iqtisodiyot fakultet
multiservis tarmoqlari
fanidan asosiy
Uzbek fanidan
mavzulari potok
asosidagi multiservis
'aliyyil a'ziym
billahil 'aliyyil
illaa billahil
quvvata illaa
falah' deganida
Kompyuter savodxonligi
bo’yicha mustaqil
'alal falah'
Hayya 'alal
'alas soloh
Hayya 'alas
mavsum boyicha


yuklab olish