Phase 12 - Lesson 12.4

Volatility Models: ARCH and GARCH

Why returns are unpredictable in the mean but not in the variance, how GARCH captures volatility clustering, and what its parameters imply.

⏱ 55 min● Advanced🔗 Prereqs: 12.1, 12.2, 11.1 MLE
↖ Phase 12 hub
Builds on: 12.1 showed zero autocorrelation of returns but strong autocorrelation of squared returns.
Leads to: GARCH volatility feeds option pricing (Phase 15) and risk models throughout.

Learning Objectives

Click a status chip to cycle: Not started → In progress → Studied → Practiced → Needs review → Mastered.

Key Vocabulary

Volatility clustering
The empirical tendency of large moves to follow large moves: \(|r_t|\) is autocorrelated even when \(r_t\) is not.
Conditional variance
\(\sigma_t^2=\Var(r_t\mid\mathcal F_{t-1})\), the variance given the past; time-varying in ARCH/GARCH.
ARCH(\(q\)
\(\sigma_t^2=\omega+\sum_{i=1}^q\alpha_i r_{t-i}^2\); conditional variance driven by recent squared returns.
GARCH(\(p,q\)
Adds lagged variances: \(\sigma_t^2=\omega+\sum\alpha_i r_{t-i}^2+\sum\beta_j\sigma_{t-j}^2\).
Persistence
\(\alpha+\beta\) in GARCH(1,1); how slowly volatility shocks decay; near 1 means very persistent.
Unconditional variance
\(\omega/(1-\alpha-\beta)\) for stationary GARCH(1,1); the long-run average variance.
IGARCH
The boundary \(\alpha+\beta=1\): shocks to variance never die out (integrated volatility).

Intuition & Motivation

Intuition
Daily returns are close to unpredictable in their mean: their autocorrelation is tiny, consistent with efficient markets. But their magnitude is highly predictable. Turbulent days cluster together and calm days cluster together - the ACF of squared or absolute returns decays slowly even though the ACF of returns is flat. ARMA cannot express this because it assumes a constant variance. We need a model where the variance itself has memory.

ARCH makes today’s variance a function of recent squared shocks: a big surprise raises tomorrow’s expected variance. GARCH adds a smoothing term - today’s variance also remembers yesterday’s variance - which captures the long, gentle decay of real volatility with very few parameters. The single number \(\alpha+\beta\) summarizes how persistent turbulence is: close to 1 (as it usually is for equities) means a volatility shock takes many weeks to fade, and exactly 1 means it never does.

Why ARMA is not enough

Write a return as \(r_t=\sigma_t z_t\) with i.i.d. standardized \(z_t\) (\(\E z_t=0,\ \Var z_t=1\)) and a time-varying conditional standard deviation \(\sigma_t\). Then \(r_t\) is serially uncorrelated (so its ACF is flat, matching returns) but \(r_t^2\) is autocorrelated (matching the clustering). The mean model and the variance model are separate: ARMA handles the mean, ARCH/GARCH the conditional variance \(\sigma_t^2\).

ARCH and GARCH defined

\[\text{ARCH}(q):\ \sigma_t^2=\omega+\sum_{i=1}^q\alpha_i r_{t-i}^2,\qquad \text{GARCH}(p,q):\ \sigma_t^2=\omega+\sum_{i=1}^q\alpha_i r_{t-i}^2+\sum_{j=1}^p\beta_j\sigma_{t-j}^2,\] (12.5)

with \(\omega\gt 0,\ \alpha_i,\beta_j\ge0\) to keep \(\sigma_t^2\gt 0\). GARCH(1,1) - just three parameters - is the workhorse and fits most return series remarkably well. Parameters are estimated by maximum likelihood: assuming \(z_t\sim\Normal(0,1)\), maximize \(\ell=-\tfrac12\sum_t\big(\log\sigma_t^2+r_t^2/\sigma_t^2\big)\) over \((\omega,\alpha,\beta)\) (a numerical optimization, since \(\sigma_t^2\) is recursive).

Worked Example - Stationarity and unconditional variance of GARCH(1,1)
1
Take expectations of \(\sigma_t^2=\omega+\alpha r_{t-1}^2+\beta\sigma_{t-1}^2\) using \(\E[r_{t-1}^2]=\E[\sigma_{t-1}^2]\) (since \(\E z^2=1\)).
2
In the stationary state \(\E[\sigma_t^2]=\bar\sigma^2\) is constant: \(\bar\sigma^2=\omega+\alpha\bar\sigma^2+\beta\bar\sigma^2\).
3
Solve: \(\bar\sigma^2=\dfrac{\omega}{1-\alpha-\beta}\), which is positive and finite iff \(\alpha+\beta\lt 1\) - the covariance-stationarity condition.
4
If \(\alpha+\beta=1\) (IGARCH) the unconditional variance is infinite - volatility shocks are permanent; if \(\gt 1\) the process is explosive.

Persistence \(\alpha+\beta\) controls the decay of a volatility shock: the conditional variance mean-reverts toward \(\bar\sigma^2\) at rate \(\alpha+\beta\) per step, giving a half-life \(\log(1/2)/\log(\alpha+\beta)\). For daily equity returns \(\alpha+\beta\) is commonly \(\approx0.98\text{--}0.99\), a half-life of weeks. GARCH also generates fat tails in the unconditional return distribution even with Gaussian \(z_t\), matching a key stylized fact; a Student-\(t\) innovation fits the tails even better.

Key Idea
Two facts to hold together: returns are (almost) unforecastable in the mean but volatility is strongly forecastable and persistent. Risk, VaR, and option prices depend on that forecastable variance - which is why GARCH is a core quant tool even though it says nothing about the direction of returns.

Interactive: simulate GARCH and see the clustering

Common Mistakes to Avoid
  • Applying ARMA to returns and assuming constant variance - it misses clustering entirely and mis-states risk.
  • Reading GARCH as a mean-return forecast; it forecasts variance, not direction.
  • Estimating GARCH without the constraints \(\omega\gt 0,\ \alpha,\beta\ge0,\ \alpha+\beta\lt 1\); violating them gives negative or explosive variances.
  • Assuming Gaussian innovations when tails are heavy; use Student-\(t\) and check standardized-residual diagnostics.
  • Ignoring leverage (down moves raise vol more than up moves); symmetric GARCH misses it - use EGARCH/GJR.
Quant Practitioner Tips
  • Fit the mean (often just a constant, or an AR(1)) and the variance jointly, or fit GARCH to the mean residuals.
  • Report persistence \(\alpha+\beta\) and the implied half-life; near-1 persistence warns that shocks fade slowly.
  • Check standardized residuals \(r_t/\hat\sigma_t\): they should be i.i.d. with no remaining ACF in their squares.
  • For asymmetry, prefer GJR-GARCH or EGARCH; for tails, Student-\(t\) innovations - both matter for equity risk.

Knowledge Check

Q1 Medium
Volatility clustering shows up statistically as:
Autocorrelated returns
Autocorrelated squared (or absolute) returns despite uncorrelated returns
A linear trend in prices
Constant conditional variance
Q2 Medium
For a covariance-stationary GARCH(1,1), the unconditional variance is:
\(\omega\)
\(\omega/(1-\alpha-\beta)\)
\(\alpha+\beta\)
\(\omega/(\alpha+\beta)\)
Q3 Medium
A GARCH(1,1) fit gives \(\alpha+\beta=0.99\). This means:
Volatility shocks decay almost instantly
Volatility is highly persistent, with a long half-life
The mean return is 0.99
The model is misspecified

Practical Exercise

A GARCH(1,1) fit to daily equity returns yields \(\omega=0.00001,\ \alpha=0.08,\ \beta=0.90\) (returns in decimals). (a) Is it stationary, and what is the unconditional daily volatility? (b) What is the persistence and the volatility half-life? (c) Interpret what happens to forecast volatility after a large market shock.

▶ Show full solution

(a) \(\alpha+\beta=0.98\lt 1\), so it is covariance-stationary. Unconditional variance \(\bar\sigma^2=\omega/(1-\alpha-\beta)=0.00001/0.02=0.0005\), so daily volatility \(\bar\sigma=\sqrt{0.0005}\approx0.0224\), about 2.24% per day (roughly \(0.0224\sqrt{252}\approx35\%\) annualized).

(b) Persistence \(\alpha+\beta=0.98\). Half-life \(=\log(0.5)/\log(0.98)\approx34\) trading days: a volatility shock takes about a month and a half to halve.

(c) A large \(r_{t-1}^2\) spikes \(\sigma_t^2\) through the \(\alpha\) term; because \(\beta=0.90\) is large, the elevated variance then feeds forward and decays only slowly toward \(\bar\sigma^2\). So after a shock the model forecasts a persistent regime of elevated volatility that mean-reverts over roughly the half-life - capturing the clustering seen in crises.

After the reveal, answer for yourself: GARCH cannot tell you where the market goes, but it tells you how violently - and that is what risk and options are priced on.

Lesson Summary

Returns are nearly unforecastable in the mean but their variance clusters and is highly forecastable. Writing \(r_t=\sigma_t z_t\), ARCH makes \(\sigma_t^2\) depend on recent squared returns and GARCH adds lagged variances, capturing the slow decay of volatility with few parameters. GARCH(1,1) is covariance-stationary iff \(\alpha+\beta\lt 1\), with unconditional variance \(\omega/(1-\alpha-\beta)\) and persistence \(\alpha+\beta\) (near 1 for equities). It generates fat tails and drives risk, VaR, and option-pricing volatility.

Formula Sheet Additions

GARCH(1,1)
\[\sigma_t^2=\omega+\alpha r_{t-1}^2+\beta\sigma_{t-1}^2\]
Persistent conditional variance; the volatility workhorse.
Unconditional variance
\[\bar\sigma^2=\omega/(1-\alpha-\beta)\]
Finite iff \(\alpha+\beta\lt 1\); persistence \(\alpha+\beta\) sets the half-life.
Error Log Checklist
  • Did I enforce \(\omega\gt 0,\ \alpha,\beta\ge0,\ \alpha+\beta\lt 1\) in estimation?
  • Did I model the mean separately and fit GARCH to residuals?
  • Are standardized residuals i.i.d. with no ACF in their squares?
  • Did I consider fat-tailed (Student-\(t\)) innovations and leverage (EGARCH/GJR)?

Retrieval Practice

Close the lesson and answer from memory before checking. This is deliberate, effortful recall - the single highest-yield study action.

▶ Show retrieval prompts & answers
Q: Why does ARMA fail on returns, and what replaces it?
A: ARMA assumes constant variance but returns show volatility clustering (autocorrelated squares); ARCH/GARCH model the time-varying conditional variance \(\sigma_t^2\).
Q: State the GARCH(1,1) stationarity condition and unconditional variance.
A: Covariance-stationary iff \(\alpha+\beta\lt 1\); unconditional variance \(\omega/(1-\alpha-\beta)\).
Q: What does persistence \(\alpha+\beta\) near 1 imply?
A: Volatility shocks decay very slowly (long half-life); at exactly 1 (IGARCH) they never die out.

Flashcards

Click to flip. These feed the site-wide spaced-repetition queue.

GARCH(1,1)
\(\sigma_t^2=\omega+\alpha r_{t-1}^2+\beta\sigma_{t-1}^2\); fit by MLE; models variance, not mean.
Stationarity/variance
Stationary iff \(\alpha+\beta\lt 1\); unconditional variance \(\omega/(1-\alpha-\beta)\).
Stylized facts
Uncorrelated returns, clustered volatility, fat tails; GARCH captures all three.

Completion Checklist

Confidence / mastery rating
Personal notes

Source References

This lesson synthesizes and paraphrases concepts from the sources below. No copyrighted text, problem sets, or solutions are reproduced. Return to the originals for full depth.