Scribd
Upload
186 views4 pages

Insurance Dataset Description2

The bank wants to selectively target existing customers to purchase a variable annuity product in order to maximize profits. An analysis data set was assembled containing demographic and banking usage information for 10,619 customers. The data set includes 48 variables describing characteristics like account balances, deposits, and credit scores. The bank aims to build a predictive model to identify the most profitable customers to receive product offers. Purchasing the annuity would generate $100 in revenue per customer for the bank, while solicitations cost $4 each to conduct.

Uploaded by

Siddhesh Tiwari
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
186 views4 pages

Insurance Dataset Description2

The bank wants to selectively target existing customers to purchase a variable annuity product in order to maximize profits. An analysis data set was assembled containing demographic and banking usage information for 10,619 customers. The data set includes 48 variables describing characteristics like account balances, deposits, and credit scores. The bank aims to build a predictive model to identify the most profitable customers to receive product offers. Purchasing the annuity would generate $100 in revenue per customer for the bank, while solicitations cost $4 each to conduct.

Uploaded by

Siddhesh Tiwari
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Insurance Data Set Description

A bank seeks to increase sales of a variable annuity product. To do this, the bank
will send product offers to existing banking customers. To maximize profits,
however, the bank wants to be selective about whom it targets. This selectivity
will be achieved by constructing a predictive model.
To achieve their analytic objective, an analysis data set was assembled. The
data set contains 10,619 records and 48 variables, assembled from several
source tables within the banks data warehouse. The source tables include the
customer master table, the transaction detail table, the product detail table, and a
third party demographic overlay table. The variables describe each customers
demographics and usage of other banking products prior to acquisition of the
variable annuity. Two of the variables are nominally scaled; the remainder are
binary or interval. A summary of the interval and binary inputs in the analysis
data set is provided below.
Number
Variable
Label
N
Missing

ACCTAGE
Age of Oldest Account
9941
678
DDA
Checking Account
10619
0
DDABAL
Checking Balance
10619
0
DEP
Checking Deposits
10619
0
DEPAMT
Amount Deposited
10619
0
CASHBK
Number Cash Back
10619
0
CHECKS
Number of Checks
10619
0
DIRDEP
Direct Deposit
10619
0
NSF
Number Insufficient Fund
10619
0
NSFAMT
Amount NSF
10619
0
PHONE
Number Telephone Banking
9286
1333
TELLER
Teller Visits
10619
0
SAV
Saving Account
10619
0
SAVBAL
Saving Balance
10619
0
ATM
ATM
10619
0
ATMAMT
ATM Withdrawal Amount
10619
0
POS
Number Point of Sale
9286
1333
POSAMT
Amount Point of Sale
9286
1333
CD
Certificate of Deposit
10619
0
CDBAL
CD Balance
10619
0

IRA
IRABAL
LOC
LOCBAL
INV
INVBAL
ILS
ILSBAL
MM
MMBAL
MMCRED
MTG
MTGBAL
CC
CCBAL
CCPURC
SDB
INCOME
HMOWN
LORES
HMVAL
AGE
CRSCORE
MOVED
INAREA

Retirement Account
IRA Balance
Line of Credit
Line of Credit Balance
Investment
Investment Balance
Installment Loan
Loan Balance
Money Market
Money Market Balance
Money Market Credits
Mortgage
Mortgage Balance
Credit Card
Credit Card Balance
Credit Card Purchases
Safety Deposit Box
Income
Owns Home
Length of Residence
Home Value
Age
Credit Score
Recent Address Change
Local Address

10619
10619
10619
10619
9286
9286
10619
10619
10619
10619
10619
10619
10619
9286
9286
9286
10619
8683
8774
8683
8683
8478
10373
10619
10619

0
0
0
0
1333
1333
0
0
0
0
0
0
0
1333
1333
1333
0
1936
1845
1936
1936
2141
246
0
0

About half of the variables have some missing values. Many of the variables,
especially those relating to monetary amounts, have an extremely large range
and highly skewed distribution.
A summary of the nominal variables and the target variable (INS) is provided by
the FREQ procedure.
Insurance Product
Cumulative
Cumulative
INS
Frequency
Percent
Frequency
Percent

0
6959
65.53
6959
65.53
1
3660
34.47
10619
100.00

Branch of Bank
Cumulative
Cumulative
BRANCH
Frequency
Percent
Frequency
Percent

B1
922
8.68
922
8.68
B10
98
0.92
1020
9.61
B11
74
0.70
1094
10.30
B12
178
1.68
1272
11.98
B13
184
1.73
1456
13.71
B14
336
3.16
1792
16.88
B15
708
6.67
2500
23.54
B16
494
4.65
2994
28.19
B17
259
2.44
3253
30.63
B18
196
1.85
3449
32.48
B19
93
0.88
3542
33.36
B2
1744
16.42
5286
49.78
B3
920
8.66
6206
58.44
B4
1876
17.67
8082
76.11
B5
932
8.78
9014
84.89
B6
480
4.52
9494
89.41
B7
476
4.48
9970
93.89
B8
461
4.34
10431
98.23
B9
188
1.77
10619
100.00

Area Classification
Cumulative
Cumulative
RES
Frequency
Percent
Frequency
Percent

R
2672
25.16
2672
25.16
S
3753
35.34
6425
60.50
U
4194
39.50
10619
100.00

The BRANCH variable, a nominal input with 19 distinct levels, indicates the
branch in which the customers initial account was opened. The RES variable, a
nominal input with three distinct levels, classifies the customers primary
residence as rural, suburban, and urban.
The target variable for this analysis, INS, indicates acquisition of the variable
annuity over a fixed period of time. While overall acquisition rate is about 2%,
the acquisition rate in the raw analysis data is more than 34%. This reflects the
separate sampling used to generate the raw data.

The bank expects to realize an average short-term revenue of about $100 from
each customer who purchase the annuity product. It is expected to cost the bank
about $4 per solicitation (which involves an initial mail solicitation with a
telephone follow-up) to carry out the campaign.

You might also like