Analytics in Banking - acceptance of Personal loan

LOGISTIC REGRESSION
A major portion of business of banks is lending. Personal loan has a major share in lending. Wouldn't it be interesting if with the given set of data and the analytics ability of the software  predict who is in need of  loan and the chance of accepting the loan? let us  look deeper into it.

Data Description:
ID	Customer ID
Age	Customer's age in completed years
Experience	#years of professional experience
Income	Annual income of the customer (Rs 000)
PinCode	Home Address pin code.
Family	Family size of the customer
CCAvg	Avg. spending on credit cards per month (Rs 000)
Education	Education Level. 1: Undergrad; 2: Graduate; 3: Advanced/Professional
Mortgage	Value of house mortgage if any. (Rs ###)
Personal Loan	Did this customer accept the personal loan offered in the last campaign?
Securities Account	Does the customer have a securities account with the bank?
CD Account	Does the customer have a Fixed  deposit (FD) account with the bank?
Online	Does the customer use internet banking facilities?
CreditCard	Does the customer use a credit card issued by WWWXXXYYY Bank?

Steps:
1.Fix the appropriate data type for the given data
2.Choose the tool which can be used to fix the solution. A typical tool choosen is the spss modeler.
Can choose the appropriate tool of your expertise.
3. A model is created with the following nodes.

4. Source tab- excel node to input the data

5. Output tab- table node to see the input data

6. Output tab- Data audit node to see the quality of data

7. The data is made to have a required partition- may be 60- 40 % one for testing and another for training.

8. A type node is connected to make a Logistic regression

Before we move on to the logistic regression, one should understand the need for it.

Linear regression fits the data given which can predict the outcome depending upon the input given.

Logistic regression : When we need to get the output as Yes / No- 0 or 1, Acceptance/ Non acceptance, we are in need of this model. A detailed difference between the models and the mathematical variation is not given at this point of time.

9. Here our objective is to know whether one will have a need / accept a PL or not  and therefore this modeling has an appropriate fit.

10. Once the modeling is done we are in need of its evaluation and find the data who are potentially in need.

11. This process is  know as lifting. We need to identify the data/ decimate it and get the  prospective list. 

12.The whole objective is to have the maximum effectiveness and attempt only the prospective clients. This reduces the time, effort and of course the money involved in campaign, attempting, meeting etc to a greater extent.

.