FRM Part 1 Quantitative Analysis: the statistical toolkit for risk

Quantitative Analysis is 20% of the exam — probability, regression, time series, and the volatility models that feed directly into VaR. Here's the full breakdown, plus a worked EWMA volatility example.

Find the gap you didn't know you had

10 questions · about 4 minutes · no account. It names the concept underneath your wrong answers, not just the topic.
Take the free FRM Part 1 diagnostic →

Practice fresh Quant questions daily

FSRS scheduling · AI tutor included
Get started →

What FRM Part 1 Quantitative Analysis actually tests

Quantitative Analysis is 20% of the Part 1 exam:

AreaWhat it covers
Probability FundamentalsConditional probability, Bayes' theorem, counting rules
Probability DistributionsNormal, lognormal, Student's t, chi-square, F, Poisson, binomial
Statistical InferenceEstimators, confidence intervals, hypothesis testing, Type I/II errors
Linear RegressionOLS assumptions, R-squared, heteroskedasticity, multicollinearity, autocorrelation
Time-Series AnalysisStationarity, unit root tests, AR/MA/ARMA processes, cointegration
Simulation MethodsMonte Carlo, variance reduction (antithetic and control variates)
Volatility ModelingEWMA, GARCH(1,1), implied volatility
Correlation & CopulasPearson, Spearman, Kendall, tail dependence, correlation breakdown in stress

Why EWMA is worth mastering cold

EWMA is the simplest of the volatility models FRM tests, but it's tested in both directions — "given yesterday's variance and today's return, compute today's variance" and "explain why a higher lambda makes the estimate less reactive to recent shocks." Confusing the direction of the decay factor is the single most common EWMA mistake.

Sample question: EWMA Volatility

Quantitative Analysis · Medium difficulty

Yesterday's EWMA variance estimate was 0.0004 (daily). Today's return was 3%. Using a decay factor (λ) of 0.94, what is today's updated EWMA variance estimate?

A. 0.000376
B. 0.00043
C. 0.0009
D. 0.0013
The correct answer is B — 0.00043.
EWMA: σ²ₜ = λ·σ²ₜ₋₁ + (1−λ)·r²ₜ = 0.94×0.0004 + 0.06×(0.03)² = 0.000376 + 0.000054 = 0.00043. Choice A stops after just the first (decayed prior variance) term — a common error that forgets to add the new-information term entirely.

The 2026 Quantitative Analysis syllabus, reading by reading

Pinnacle runs on a syllabus graph — named readings with explicit prerequisites, each one mapped against GARP's official 2026 FRM Study Guide. It is the same map the free diagnostic reasons over, not a marketing summary of it. These are the six confirmed Quantitative Analysis readings, and what each one covers:

Probability & Distributions

Probability fundamentals — sample spaces, mutually exclusive events, conditional probability, Bayes' theorem, the law of total probability — and the named distributions used in risk: uniform, Bernoulli, binomial, Poisson, normal, lognormal, Student's t, chi-square, F, and beta.

Statistical Inference

Sample moments and the population-versus-sample distinction; the properties of estimators — bias, efficiency, consistency; and hypothesis testing with confidence intervals.

Linear Regression

OLS assumptions and parameter interpretation, R-squared and the F-statistic, residual analysis — heteroskedasticity, multicollinearity, autocorrelation and Durbin-Watson — and multiple regression in risk modelling.

Time Series & Volatility

Stationarity and unit-root tests, AR, MA and ARMA processes, cointegration and forecasting; volatility modelling with realised volatility, EWMA and GARCH(1,1); and measuring returns, volatility and correlation for normally and non-normally distributed variables.

Simulation & Bootstrapping

Monte Carlo methods: generating scenarios, pseudorandom numbers, and variance reduction via antithetic variates, control variates, and importance sampling.

Machine Learning Foundations

Preparing data for machine-learning applications, distinguishing the types of model, the leading supervised models for classification and prediction, and machine learning as an alternative to traditional model-building in finance.

Every practice question in the bank is tagged to one of these readings — which is how the diagnostic can name the specific concept underneath a wrong answer, not just the topic area it sat in.

Reading names follow GARP's published 2026 FRM Study Guide, referenced for accuracy. Pinnacle is an independent adaptive learning platform. FRM® is a registered trademark of the Global Association of Risk Professionals (GARP). Pinnacle is not affiliated with, endorsed by, or connected to that organisation.

Master Quant Analysis with adaptive practice

Fresh EWMA and regression questions daily · Full explanations
Get started →

More FRM Part 1 topics