README.md

Data_Analyze-American_Express-

American Express data analysis competition-2018

Problem Statement

You have to create a list of applications in the order in which Banco should process them. With an objective to maintain healthy financials, Banco would like to process least risky applications first. Against each application, you also have to provide your prediction of default - 1 or 0, where 1 indicates a default and 0 indicates no default.

Assume:

• A resident of the city can submit only a single application form
• None of the applications submitted are fraudulent
• State any other assumptions in your final submission

Data for Analysis

Following files can be downloaded for your analysis.

1. Training_dataset.csv: This dataset contains: a Applicant level historic credit history b Performance in terms of default tagging i.e. 1 for default and 0 for no default c Application and bureau data
2. Leaderboard_dataset.csv: This data has historical applicant level data along with all the variables in the training dataset. The actual performance i.e. default tagging is not present in this data.
3. Evaluation_dataset.csv: This data has applicant level data along with all the variables in the training dataset. The actual performance i.e. default tagging is not present in this data.
4. Data_Dictionary.xlsx: This sheet will give you the description of all the variables contained in the 3 datasets above.