Skip to content

QiangUM/CreditCardRisk

Repository files navigation

Credit-risk-analysis


In order to avoid the loss caused by defaulting on credit cards bill payment, the risk management on credit card default has been acting an imporant role in bank. Through the analysis of the user's personal profile and historical bills, we want find the reason why clients defalut on their July Statement and find the best model for predicting users' July Payment Status.

1. Data Cleaning and Redefine Value Labels

Alt text Alt text Alt text Alt text Alt text

2. Data Standardization

Alt text

3. Data Analysis

Alt text

Alt text Alt test Alt text Alt text Alt text

4. Modeling

Alt text Alt text Alt text Alt text Alt text Alt text

5. Conclusion

Among the algorithms used, K-Nearest Neigbor (k=9) gave a higher accuracy, which about 82%. However, the model only uses high-correlation variables to train the model, we still can improve our model through other feature variables.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published