Investment is an activity to hand over capital to an agency, usually a company, in the hope of getting a profitable return. The probability of profit or loss that an investor will get depends on the company being invested. Companies that have good performance will provide benefits for investors. However, companies that have poor performance have the potential to go bankrupt and will cause losses to investors. It would be better for investors, be they individuals or companies that will invest, to carry out an analysis regarding the performance of the intended company. The prediction model will also be very helpful in predicting whether a company will go bankrupt or not with financial information as a predictive model variable. This project seeks to create a predictive model with a tree-based model to help investors invest so they don't invest in companies that will go bankrupt.
The purpose of this project is to predict whether a company will go bankrupt or not by using financial variables. This project also conducts exploratory data analysis and feature engineering and performs testing using test data. What we will do in this project is:
- Perform exploratory data analysis and feature engineering
- Creating a random forest model for bankruptcy prediction
- Evaluate and improve model performance using cross validation
- perform hyperparameter tunning to find the best parameters for the model
This project consists of 3 parts:
- Prepare Data
- Imported
- Exploratory Data Analysis (EDA)
- Feature Engineering 2.Build Model
- Iterate
- Evaluate
- Communicating Results
https://scikit-learn.org/stable/modules/cross_validation.html#k-fold
https://www.kaggle.com/datasets/fedesoriano/company-bankruptcy-prediction