Lab4 277
This document was uploaded by user and they confirmed that they have the permission to share it. If you are author or own the copyright of this book, please report to us by using this DMCA report form. Report DMCA
Overview
Download & View Lab4 277 as PDF for free.
More details
- Words: 413
- Pages: 3
INTEROFFICE ME MORANDU M
TO:
PROFESSOR MCCLINTOCK
FROM:
MEGAN LEE
SUBJECT:
CHAPTER 17 LAB
DATE:
NOVEMBER 14 2018
Introduction In this document, models were ran to predict both credit card limit and the likelihood of someone defaulting on a loan. Specifically, we ran two regression models with dummy variables for predicting limit as well as likelihood of someone defaulting on a loan based on various variables. Data Analysis
Regression equation: Μ πΆπππππ‘ πΏππππ‘ = β22.94 + 302.61(ππππππ) + 921.43(π’πππ£πππ ππ‘π¦) + 3237.61(ππππ π πβπππ) + 2930.03(πππππππ) β 57.14(πππππππ β πππ) + 101.50(πππ) Interpretation of R-squared: We are 10.91% of the way to perfectly predicting credit limit using the variables in the model. Interpretation of coefficients: Since all of the variables have low p-values, we can conclude that all explanatory variables can be used. Females will have a $302 higher credit limit than men, on average all else constant. As age increases by 1 for single people, credit limit increases by $101.50, on average all else constant. As age increases by 1 for married people, credit limit increases by $44.36, on average all else constant. As age increase by 1 for single people, credit limit increases $57.14 more than for a married person. Predict limit of a 35-year-old, married, female with a university degree: -22.94 + 302.61(1) + 921.43(1) + 3237.61(0) + 2930.03(1) β 57.14(35*1) +101.50(35) = $5683.73
ππππππβπππ Μ ππ πππππ’ππ‘πππ = .208+0.00000448(Average bill amount) β 0.00014(average previous payment amount + 0.000831(age)
2
Interpretation of R-squared: We are 11.25% of the way to perfectly predicting someoneβs probability of defaulting on a payment using the variables in the model. Interpretations of coefficients: As average bill amount increases by $100, the chance of defaulting increase by .0448 percentage points, on average all else constant. As age increases by 1, likelihood of defaulting increases by .0831 percentage points, on average all else constant. Predict the chance of someone defaulting who is 40 years old, has an average bill amount of $1500, and average payments of $700: .208 + 0.00000448(1500) β 0.00014(700) +0.000831(40) = 0.14996 There is a 15% chance that someone who is 40 years old with an average bill amount of $1500 and average previous payments of $700 will default on their payment. Conclusion Two regression models were ran two with dummy variables for predicting limit as well as likelihood of someone defaulting on a loan based on various variables. Because of both of their low R-square, I would not suggest using either of these models.
3
TO:
PROFESSOR MCCLINTOCK
FROM:
MEGAN LEE
SUBJECT:
CHAPTER 17 LAB
DATE:
NOVEMBER 14 2018
Introduction In this document, models were ran to predict both credit card limit and the likelihood of someone defaulting on a loan. Specifically, we ran two regression models with dummy variables for predicting limit as well as likelihood of someone defaulting on a loan based on various variables. Data Analysis
Regression equation: Μ πΆπππππ‘ πΏππππ‘ = β22.94 + 302.61(ππππππ) + 921.43(π’πππ£πππ ππ‘π¦) + 3237.61(ππππ π πβπππ) + 2930.03(πππππππ) β 57.14(πππππππ β πππ) + 101.50(πππ) Interpretation of R-squared: We are 10.91% of the way to perfectly predicting credit limit using the variables in the model. Interpretation of coefficients: Since all of the variables have low p-values, we can conclude that all explanatory variables can be used. Females will have a $302 higher credit limit than men, on average all else constant. As age increases by 1 for single people, credit limit increases by $101.50, on average all else constant. As age increases by 1 for married people, credit limit increases by $44.36, on average all else constant. As age increase by 1 for single people, credit limit increases $57.14 more than for a married person. Predict limit of a 35-year-old, married, female with a university degree: -22.94 + 302.61(1) + 921.43(1) + 3237.61(0) + 2930.03(1) β 57.14(35*1) +101.50(35) = $5683.73
ππππππβπππ Μ ππ πππππ’ππ‘πππ = .208+0.00000448(Average bill amount) β 0.00014(average previous payment amount + 0.000831(age)
2
Interpretation of R-squared: We are 11.25% of the way to perfectly predicting someoneβs probability of defaulting on a payment using the variables in the model. Interpretations of coefficients: As average bill amount increases by $100, the chance of defaulting increase by .0448 percentage points, on average all else constant. As age increases by 1, likelihood of defaulting increases by .0831 percentage points, on average all else constant. Predict the chance of someone defaulting who is 40 years old, has an average bill amount of $1500, and average payments of $700: .208 + 0.00000448(1500) β 0.00014(700) +0.000831(40) = 0.14996 There is a 15% chance that someone who is 40 years old with an average bill amount of $1500 and average previous payments of $700 will default on their payment. Conclusion Two regression models were ran two with dummy variables for predicting limit as well as likelihood of someone defaulting on a loan based on various variables. Because of both of their low R-square, I would not suggest using either of these models.
3
