Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit

LUBS2226

Applied Credit Analytics

Part A: Answer at least one question from this section. Mark allocations are provided for questions that have multiple parts.

Question 1. A market research data set contains survey responses from 856 people who have either tried or  not tried a company’s  new frozen  lasagne  product. The company  is working  out  how  best  to  market  the  new  product.  The  file  contains  a  number  of demographics on these customers, their age, weight, income in ($), pay type, car value ($), credit card debt ($), gender, whether they live alone, dwelling type, monthly number of trips to the shopping centre, and neighbourhood type.

The   categorical   dependent  variable,   Have  Tried,   indicates  the   customers  who   have purchased the product. The company wants to understand why some potential customers are triers and others are not. What distinguishes triers from non-triers?

Table 1 provides the analysts initial attempt at estimating a linear probability model (LPM) using Ordinary Least Squares (OLS) and Table 2 provides results from the estimation of a    logistic regression using the same variables.

(a) Provide an interpretation of the Linear Probability Model, LPM (Table 1). Including a   discussion of the choice of the independent variables and their signs and significance.

(50 Marks)

(b) Provide an interpretation of the Logit Model (Table 2). Including a discussion of the

signs and significance of the independent variables and the overall classification

accuracy of the model. (25 Marks)

(c) Discuss the advantages and disadvantages of using the LPM regression in comparison to

the Logistic Regression i.e. what are the strengths and weaknesses of each approach.

(25 Marks) Total 100 Marks


Table 1

Linear Probability Model

Model

Coefficient B Standard error

T-statistic

(Constant)

0.141

0.124

1.136

Age

-0.010

0.001

-7.446

weight in lbs

0.001

0.001

2.341

Income

0.000

0.000

1.572

Car Value

0.000

0.000

-1.097

Credit Card Debt

0.000

0.000

1.148

Number of Shopping Trips 0.087

0.006

14.890

Pay Type Salaried =1 0.180

0.030

6.006

Gender Male=1

0.039

0.026

1.512

Live Alone Yes=1

0.165

0.035

4.642

Home Type Appartment=1 -0.013

0.036

-0.348

Home Type House=1 0.020

0.033

0.605

Dependent Variable: Tried Yes=1; Tried No=0


Sum of Squares

df

Mean Square

Regression

88.399

11

8.036

Residual

120.357

844

0.143

Total

208.756

855