Regime Switching & GARCH

Financial markets do not behave the same way every day. Periods of calm can be followed by periods of extreme volatility. GARCH and regime-switching models give us ways to describe this changing behaviour.

01   Volatility Changes Through Time

We have already seen that financial returns can have fat tails and skewness. But there is another important feature: volatility itself changes over time.

Markets can spend weeks or months in relatively calm conditions and then suddenly experience a period of large daily movements.

This creates a phenomenon known as volatility clustering.

Large movements tend to be followed by large movements
Small movements tend to be followed by small movements
The important point is that volatility is not necessarily constant. Knowing what happened recently can tell us something about the volatility we might expect tomorrow.

02   Seeing Volatility Clustering

The chart below simulates returns where volatility moves between periods of calm and periods of turbulence.

Notice how the size of the movements tends to cluster together. There are relatively quiet periods followed by groups of much larger movements.

03   The GARCH Idea

GARCH stands for Generalized Autoregressive Conditional Heteroskedasticity.

The name is complicated, but the underlying idea is relatively simple: today's volatility depends partly on recent volatility and recent returns.

A basic GARCH(1,1) model can be written as:

σ2t = ω + αε2t-1 + βσ2t-1

The three components have intuitive interpretations.

ω

The long-run baseline level of variance in the model.

α

How strongly volatility reacts to a recent shock.

β

How persistent volatility is once it has changed.

ε

The unexpected return or shock observed in the previous period.

04   Explore a GARCH Process

The following simulation allows us to change the parameters controlling how quickly volatility reacts to shocks and how persistent it is.

0.12
0.82
α
0.12
β
0.82
α + β
0.94
Behaviour
Persistent
When α + β is close to 1, shocks to volatility tend to persist for a long time. This is one of the key features that makes GARCH useful for financial time series.

05   Volatility Forecasting

One reason GARCH models are useful is that they allow us to estimate conditional volatility.

Instead of assuming that tomorrow's volatility is always the same, the model can respond to what has happened recently.

Recent shocks  →  Current volatility  →  Future volatility

06   GARCH Still Has a Limitation

GARCH is very good at modelling changing volatility, but it does not necessarily tell us why volatility has changed.

Imagine that markets move between fundamentally different environments:

Calm Regime

Lower volatility, smaller daily movements and relatively stable market conditions.

Stress Regime

Higher volatility, larger movements and much greater uncertainty.

This leads naturally to the idea of regime switching.

07   What Is Regime Switching?

A regime-switching model assumes that the market can exist in one of several underlying states, or regimes.

We do not necessarily observe the regime directly. Instead, we observe returns and infer which regime is most likely.

Regime 1  ↔  Regime 2  ↔  Regime 3

The model also estimates the probability of moving from one regime to another.

Think of the regimes as different "states of the market". A calm market is more likely to remain calm tomorrow, but there is always some probability of switching into a more volatile state.

08   A Two-Regime Market

Here we simulate a market that switches between a low-volatility and high-volatility regime.

92%

Low Volatility

0.5%

Typical daily volatility in the simulated calm regime.

High Volatility

2.0%

Typical daily volatility in the simulated stress regime.

Persistence

92%

Probability of remaining in the current regime.

09   GARCH + Regime Switching

The two approaches solve related but different problems.

GARCH

Dynamic volatility

Volatility responds continuously to recent shocks and gradually returns toward its long-run level.

Regime Switching

Different states

The market can move between distinct underlying environments.

Combined

Both

Volatility can evolve dynamically while also having different target levels in different regimes.

Regime  →  Target Volatility  →  GARCH Dynamics  →  Daily Return

10   A Simple Three-Regime Model

We can extend the idea to three states.

Regime 1

Calm

Low target variance and relatively small movements.

Regime 2

Normal

Intermediate volatility and more typical market conditions.

Regime 3

Stress

High target variance and much larger potential movements.

11   Why This Matters for the Pension Model

This is where the concepts become directly relevant to our financial modelling work.

A simple Monte Carlo model might assume that investment returns have constant volatility and come from one distribution.

A more realistic model can allow:

Fitted Distribution  +  GARCH  +  Regime Switching  →  Realistic Return Simulation
This is considerably more sophisticated than simply drawing independent Normal returns. It allows the simulated investment paths to experience periods of calm, periods of stress and persistent changes in volatility.

12   From Mathematical Beauty to Financial Modelling

We have now travelled quite a long way from the original idea of probability.

Probability → Distributions → Sampling → Monte Carlo → Financial Returns → Student-t / Skew-t → GARCH → Regime Switching

Each step adds another layer of realism.

The important lesson is that the mathematics is not just about complicated formulas. Each model is trying to capture something we can actually observe in the real world.

The Connection to KerryTech

These ideas provide the mathematical foundation for the more advanced financial simulations elsewhere on the site.

The next step is to take these concepts and apply them to an actual fitted financial return model.

← Back to Student-t & Skew-t

← Back to KerryTech