Overfitting in Financial Model Building

Creating a powerful predictive algorithm usually involves a certain amount of hyperparameter optimization. This involves tuning a model’s parameters to maximize a certain objective function, such as the Sharpe Ratio in finance. One of the most popular methods is Bayesian optimization, which is a significant improvement in computational efficiency and results over both random search and grid search — two other popular ways of optimizing hyperparameters. When evaluating a costly black-box function, Bayesian optimization is by far the most popular method for tuning hyperparameters.

Read the full story

You may also like...