About this question

Bias-Variance Decomposition

Medium · Linear Algebra & Machine Learning · Quant Trader interview question · machine-learning, bias-variance, model-complexity, statistics

You are building a model to predict the next-day return of a specific stock. You know that the expected prediction error (Mean Squared Error, or MSE) can be decomposed into three components: bias squared, variance, and irreducible error. How does increasing model complexity (e.g., using higher-order polynomial features in a linear regression, or increasing the depth of a decision tree) generally affect these three components?