CFA Level II ExamQuantitative MethodsEasy
A portfolio manager is evaluating a new machine learning algorithm designed to predict stock prices. After extensive training, the algorithm achieves a 99% accuracy on the training data but performs poorly on new, unseen market data. This situation is most indicative of:
- AOverfitting
- BUnderfitting
- CLack of feature engineering
- DBias-variance trade-off optimization
Show answer & explanationAnswer & explanation
Correct answer: A. Overfitting
Overfitting occurs when a model learns the training data too well, including its noise and specific patterns, leading to excellent performance on training data but poor generalization to new, unseen data.
Why the other options are wrong
- B. Underfitting occurs when a model is too simple to capture the underlying patterns in the data, performing poorly on both training and test data.
- C. Lack of feature engineering might lead to underfitting or generally poor performance, but not specifically to high training accuracy coupled with low test accuracy.
- D. Bias-variance trade-off optimization is a goal in model building, not a description of this specific outcome; overfitting is a result of one side of this trade-off.
Overfitting
A modeling error that occurs when a function is too closely aligned to a limited set of data points, resulting in poor performance on new, unseen data.
- High accuracy on training data.
- Low accuracy on validation/test data.
- Model learns noise and specific patterns of training data.
Memory trick: Training too hard makes a model too specific.