Using the data set in excel, suggest the best regression model to explain the cr

Question

Using the data set in excel, suggest the best regression model to explain the credit approval decisions

Credit Approval Decisions Homeowner Credit Score Years of Credit History Revolving Balance Revolving Utilization Decision Y 725 20 $11,320 25% Approve Y 573 9 $7,200 70% Reject Y 677 11 $20,000 55% Approve N 625 15 $12,800 65% Reject N 527 12 $5,700 75% Reject Y 795 22 $9,000 12% Approve N 733 7 $35,200 20% Approve N 620 5 $22,800 62% Reject Y 591 17 $16,500 50% Reject Y 660 24 $9,200 35% Approve Y 700 19 $22,000 18% Approve Y 500 16 $12,500 83% Reject Y 565 6 $7,700 70% Reject N 620 3 $37,400 87% Reject Y 774 13 $6,100 7% Approve Y 802 10 $10,500 5% Approve N 640 7 $17,300 59% Reject N 523 14 $27,000 79% Reject Y 811 20 $13,400 3% Approve N 763 2 $11,200 70% Reject N 555 4 $2,500 100% Reject N 617 9 $8,400 34% Reject Y 642 13 $16,000 25% Approve N 688 3 $3,300 11% Approve Y 649 12 $7,500 5% Approve Y 695 15 $20,300 22% Approve Y 701 9 $11,700 15% Approve N 635 7 $29,100 85% Reject N 507 2 $2,000 100% Reject Y 677 12 $7,600 9% Approve N 485 5 $1,000 80% Reject N 582 3 $8,500 65% Reject Y 699 17 $12,800 27% Approve Y 703 22 $10,000 20% Approve N 585 18 $31,000 78% Reject Y 620 8 $16,200 55% Reject Y 695 16 $9,700 11% Approve Y 774 13 $6,100 7% Approve Y 802 10 $10,500 5% Approve N 640 7 $17,300 59% Reject N 536 14 $27,000 79% Reject Y 801 20 $13,400 3% Approve N 760 2 $11,200 70% Reject N 567 4 $2,200 95% Reject N 600 10 $12,050 81% Reject Y 702 11 $11,700 15% Approve Y 636 8 $29,100 85% Reject N 509 3 $2,000 100% Reject N 595 18 $29,000 78% Reject Y 733 15 $13,000 24% Approve

Explanation / Answer

Using whole variables

a. R Squared = .690 (Adjusted R Squared = .647)

Using variables which is significant upto 0.10 level of significance.

a. R Squared = .665 (Adjusted R Squared = .643)

Again removing those  variables which is not significant at 0.10 level of significance

a. R Squared = .650 (Adjusted R Squared = .635)

Similarly

a. R Squared = .634(Adjusted R Squared = .627)

From the above models, we can say that the last model is the best because even in reducing the explanaotory variables the R-square value does not reduce significantly.

Tests of Between-Subjects Effects Dependent Variable:Credit score Source Type III Sum of Squares df Mean Square F Sig. Corrected Model 273987.851a 6 45664.642 15.947 .000 Intercept 1.507E+06 1 1.507E+06 526.391 .000 Year 9864.697 1 9864.697 3.445 .070 Revolving_Balance 12675.967 1 12675.967 4.427 .041 Revolving_Utilization 45969.292 1 45969.292 16.054 .000 Homeowner 1843.672 1 1843.672 .644 .427 Decision 78.491 1 78.491 .027 .869 Homeowner * Decision 8216.510 1 8216.510 2.869 .098 Error 123130.229 43 2863.494 Total 2.167E+07 50 Corrected Total 397118.080 49

Related Questions

Navigate

• Browse (All)
• Browse U
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.