Unsupervised ML Algorithms
If he’d had ML algorithms, he may have fared better…
Trading stock is highly competitive and carries a massive amount of risk to the uninformed investor. Without an MBA, it can be difficult to wade through the sea of financial data available for firms traded on the open market. Furthermore, who’s to say companies aren’t cooking the books anyway? With Machine Learning, we can delve deep into the unseen relationships between performance metrics and reliably group them based on their fundamentals. The algorithm was able to reliably separate the companies into 5 distinct groups with very different business dynamics and likelihoods of providing an ROI. It appears to have also identified a group of shady companies that may be misrepresenting their numbers at the shareholders expense.
The data was cleaned, formatted and scaled.
EDA.
K-Means Clustering with Silhouette Scoring.
Hierarchical Clustering.
PCA and t-SNE for dimensionality reduction and visualization.
