Heteroskedasticity

What it means

Regression errors whose variance is not constant across observations.

The exam angle

Coefficients stay unbiased; the standard errors do not. So t-statistics are wrong and inference is unreliable — the fix is robust standard errors, not a different slope.

Where it appears in the syllabus

• Quantitative MethodsCFA Level II · exam weight 5–10% · Multiple regression, misspecification, time series, machine learning.
• Quantitative AnalysisFRM Part I · exam weight 20% · Probability, hypothesis testing, regression, time series, volatility, simulation.

Related terms

• Coefficient of determinationThe fraction of the dependent variable’s variation explained by the regression.
• Linear regressionFitting a straight line through data by minimising the sum of squared residuals, to explain a dependent variable by…
• MulticollinearityTwo or more independent variables in a regression that are highly correlated.
• Serial correlationRegression errors correlated with their own past values.
• OverfittingA model that fits the noise in its training data as well as the signal, so it performs well in-sample and poorly on…
• Supervised learningMachine learning that fits a model to examples with known outcomes — labelled data — so it can predict the outcome for…