Overfitting
A model learning the noise in its training data as though it were signal. It fits the past closely and predicts badly, which is why models are tested on data held out from training.
Where you will meet it
Overfitting appears in 7 learning outcomes on the CFA syllabus. A few of them:
- Explain what is meant by machine learning, artificial intelligence and big data
“The recurring failure is overfitting, where a model learns the training data's noise as though it were signal, which is why a model is judged on data it has never seen.”
- Explain overfitting and recognise ways to deal with it
“Total error splits into bias error from underfitting, variance error from overfitting, and irreducible base error.”
- Set out the steps by which a quantitative active equity strategy is developed
“Pitfalls include data-mining and overfitting, back-tests that understate turnover and costs, signal decay as others crowd in, and liquidity squeezes when many quant funds unwind similar positions.”
- Explain deep, supervised and unsupervised learning and how investment managers use them
“The recurring danger is overfitting: a model that fits its training data closely but performs poorly on new data.”
Practise it
Practise overfitting until it is automatic: questions that explain every wrong option.
Take the free diagnostic →