We live in a time where decentralized finance has been on the rise and more businesses have been closing than ever before. My last post analyzed posts from a group dedicated to the downfall of rich hedge fund investors. Now I will be doing a different type of analysis.
I created a model to predict the likelihood of a business going bankrupt in Taiwan. My data set came from Kaggle and fortunately has no errors. After ensuring that the data had no issues, I began developing the model. I used a handy method from SKLearn called “feature selection”. It algorithmically selects the most useful features out of a set of data. I chose to stick with the top 50 features for my model to train from. After scaling the data to a standard format, I conducted a test-train split, a logistic regression, and implemented a classification report to verify results. The machine yielded adequate results suggesting that it can predict with appropriate accuracy the likelihood of a company going sinking to the bottom of the financial sea.