Scribd
Upload
124 views3 pages

Lending Club Data Analysis PDF

This document outlines a project for analyzing lending data from Lending Club to predict borrower default rates. The objectives are to clean and explore the data, engineer features, predict default rates using classification and regression algorithms, compare model performance, and write a final report. The data includes over 100 attributes on loans issued by Lending Club. Students will form teams to complete the milestones of data preparation, analysis, feature engineering, modeling, evaluation, and reporting. References are provided for related previous work.

Uploaded by

azeezvidhya
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
124 views3 pages

Lending Club Data Analysis PDF

This document outlines a project for analyzing lending data from Lending Club to predict borrower default rates. The objectives are to clean and explore the data, engineer features, predict default rates using classification and regression algorithms, compare model performance, and write a final report. The data includes over 100 attributes on loans issued by Lending Club. Students will form teams to complete the milestones of data preparation, analysis, feature engineering, modeling, evaluation, and reporting. References are provided for related previous work.

Uploaded by

azeezvidhya
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/ 3

HARVARD UNIVERSITY CS109

Lending Club Data Analysis and Default


Rate Prediction
Proposed by Qing Zhao

Lending Club (LC) is the world’s largest online marketplace connecting borrowers and investors. It is
transforming the banking system to make credit more affordable and investing more rewarding. Lending Club
operates at a lower cost than traditional bank lending programs and pass the savings on to borrowers in the
form of lower rates and to investors in the form of solid risk-adjusted returns.

Details of how it works can be found here.

OBJECTIVE
In this project, you are expected to play with the data provided by LC, conduct a set of exploratory analysis
and try to apply various machine learning techniques to predict borrower’s default rate.

DATA
All data regarding this project can be accessed here. So far the data is available till 2016 Q2 and can be dated
back to 2007. The data consists in 4 files updated every quarter on the same day as the quarterly results of
the company are released. They contain information on almost all the loans issued by LC. The only loans
missing from these files are the few loans where LC was not authorized to release publicly the details of the
transactions.

The information available for each loan consists of all the details of the loans at the time of their issuance as
well as more information relative to the latest status of loan such as how much principal has been paid so far,
how much interest, if the loan was fully paid or defaulted, or if the borrower is late on payments etc.

LOAN DATA

These files contain complete loan data for all loans issued through the time period stated, including the
current loan status (Current, Late, Fully Paid, etc.) and latest payment information. The file containing
loan data through the "present" contains complete loan data for all loans issued through the previous
completed calendar quarter. You need to register an LC account and sign in to download the full
version of the files.

• DECLINED LOAN DATA


These files contain the list and details of all loan applications that did not meet Lending Club's credit
underwriting policy.
HARVARD UNIVERSITY CS109

There are 115 data attributes in total, however, not all of them can contribute to data analysis or can be used
as features for machine learning techniques. It is your responsibility to identify which features to use, or even
better, create your own features based on the given attributes.

MILESTONES
This part serves as a suggested project planning. Some parts can be later adjusted through discussions with
TF.

1. Project Selection

Form teams of 2 or 3 and select a project from the provided list.

2. Data Cleaning and Exploratory Analysis

The data provided in this project is relatively clean and formatted, however, you may need to deal with
some null values or transform some categorical variables. You are also encourage to seek for external
data sources.

Based on the preprocessed data, you are expected to conduct a series of exploratory data analysis to
get a taste of how the data looks like or how it is distributed. In this part you are required to
perform various data visualisation to show what you have found.

3. Feature Engineering

Since there more over 100 data fields (115) provided, it is not necessary, even not suggested to use all
of them. It is your job to perform feature selection and discover the features that are indicative of
someone paying or defaulting on their loan.

4. Predict Default Rate

Here comes the most exciting part! Based on the previous data exploration and feature engineering,
you can predict whether someone will default based on some classification algorithms or predict
someone’s default possibility using various regressions so as to avoid loans that are predicted to
default.

Possible algorithms include but not limited to: Logistic Regression, SVM, Naive Bayes, Linear
Regression, Random Forest, Ada Boost, etc.

5. Model Comparison and Evaluation

Here you are expected to come up with a set of evaluation criteria to compare different model’s
performance and discuss their “Bias vs Variance”. It is important to try to explain why the models give
such results and what are their pros & cons under certain constraints.

6. Final Report
HARVARD UNIVERSITY CS109

As the last step, please write a final report in iPython notebook to include all your codes, graphs and
writeup.

REFERENCE
Here are some works done previously on this topic, feel free to take a look and grab things you think could be
helpful to your work. You are also encouraged to research more on related work online. However, you MUST
come up with your own ideas and mark all your references at the end of your final report.

• https://www.lendingclub.com/info/statistics.action

• http://www.lendingmemo.com/lending-club-prosper-default-rates/

• http://res.cloudinary.com/general-assembly-profiles/image/upload/v1416535475/
uwumoooppttsmpgu1goo.pdf

• http://nbviewer.jupyter.org/gist/odubno/0b767a47f75adb382246

You might also like