Building a Logistic Regression Model in SAS EM
October 27, 2003
The Data and the Task
We will use the same data set,
myraw.xls
, as was used in a few earlier recitations. Recall
that this data set was extracted from a larger data set maintained by a national veteran’s
organization on people who have made charitable donations to them in the past.
The
organization has just finished a new fundraising campaign specifically targeting previous
donors, and they want an analysis of the results. In one of the earlier recitations we used
a linear regression model to predict the amount a donor contributed (assuming that the
donor did contribute to the campaign) based on database information about that donor. In
this recitation, we will use the data to predict
whether or not
someone donated to the most
recent campaign using a logistic regression model.
As before, the data set is an Excel spreadsheet for which each row represents a past donor
(an observation) and each column represents a category of information about the donor (a
variable). The meanings of the variables are reprinted below:
Variable Name
Role
Type
Description
AGE
Input
Interval
Age
AVGGIFT
Input
Interval
Average dollar amount of gifts to date
CARDGIFT
Input
Interval
Number of lifetime gifts to card promotions
CARDPROM
Input
Interval
Lifetime number of card promotions received
FEDGOV
Input
Interval
% of household employed by federal government
FIRSTT
Input
Interval
Elapsed time since first donation
GENDER
Input
Binary
F = female, M = male
HOMEOWNR
Input
Binary
H = homeowner, U = unknown
IDCODE
ID
Nominal
ID code, unique for each person
INCOME
Input
Ordinal
Household income level (integers 09)
LASTT
Input
Interval
Elapsed time since last donation
LOCALGOV
Input
Interval
% of household in local government
MALEMILI
Input
Interval
% of household males active in the military
MALEVET
Input
Interval
% of household male veterans
NUMPROM
Input
Interval
Lifetime number of promotions received to date
PCOWNERS
Input
Binary
Y = donor owns computer (missing otherwise)
PETS
Input
Binary
Y = donor owns pets (missing otherwise)
STATEGOV
Input
Interval
% of household in state government
TARGET
B
Target
Binary
1 = gave to new campaign, 0 = did not
TARGET
D
Target
Interval
Dollar amount of contribution to new campaign
TIMELAG
Input
Interval
Number of months between first and second gift
The data set contains 2,988 observations, and we will make use of all of them.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
This is the end of the preview.
Sign up
to
access the rest of the document.
 Spring '07
 APANASOVICH
 Regression Analysis, logistic regression model, SAS EM, Data Source node, Input Interval

Click to edit the document details