BostonMortgages: Mortgage applications in Boston

In HRW: Datasets, Functions and Scripts for Semiparametric Regression Supporting Harezlak, Ruppert & Wand (2018)

Description

The full dataset is in the data frame 'Hdma' within the R package Ecdat. This data frame is the subset of mortgage applications during the years 1998-1999.

Usage

data(BostonMortgages)

Format

A data frame with 2,380 observations on the following 13 variables:

dir

ratio of the debt payments to the total income.

hir

ratio of the housing expenses to the total income.

lvr

ratio of the loan size to the assessed value of property.

ccs

a credit score ranging from 1 to 6, where a low value indicates low credit risk.

mcs

a mortgage credit score from 1 to 6, where a low value indicates low credit risk.

pbcr

did the applicant have a public bad credit record?: a factor with levels no and yes.

dmi

was the applicant denied mortgage insurance?: a factor with levels no and yes.

self

was the applicant self-employed?: a factor with levels no and yes.

single

is the applicant single?: a factor with levels no and yes.

uria

1989 Massachusetts unemployment rate in the applicant's industry.

condominium

is the unit a condominium?: a factor with levels no and yes.

black

is the applicant black?: a factor with levels no and yes.

deny

was the mortgage denied?: a factor with levels no and yes.

References

Munnell, A. H., Tootell, G. M. B., Browne, L. E., and McEneaney, J. (1996). Mortgage lending in Boston: Interpreting HDMA data, American Economic Review, 25-53.

Examples

library(HRW) ; data(BostonMortgages)
BostonMortgages$denyBinary <- as.numeric(BostonMortgages$deny == "yes")
fit <- glm(denyBinary ~ black + dir + lvr + pbcr + self + single + as.factor(ccs),
                        family = binomial,data = BostonMortgages)
summary(fit)

HRW documentation built on Nov. 23, 2021, 9:07 a.m.