What it means
Before Engle, many models assumed that the size of market swings was constant. Anyone watching markets knows that is wrong: quiet spells are followed by quiet spells and turbulent days are followed by more turbulence.
Engle introduced the autoregressive conditional heteroskedasticity model, usually shortened to ARCH, to capture this pattern. In plain English, ARCH says that today's expected volatility depends on the size of recent surprises.
If yesterday's price move was large, today's variance (the squared measure of spread) is predicted to be higher. This simple link between past shocks and future risk turned out to describe real markets remarkably well.
His student Tim Bollerslev later generalised the idea into GARCH, which also includes the previous variance forecast, and countless variations followed. GARCH-type models are now standard tools in banks, hedge funds and regulators for estimating value at risk, pricing options and setting margin requirements.
Engle shared the 2003 prize with Clive Granger, who was recognised for work on cointegration. Engle later worked at New York University's Stern School of Business, where he also helped build tools to measure systemic risk, the danger that distress at one institution spreads through the whole financial system.
The nuance is that these models forecast the size of moves, not their direction. They help firms judge how risky tomorrow is likely to be, but they cannot tell anyone whether the market will go up or down.
For managers, the practical message is that risk estimates must be refreshed. A risk number calculated last year, or even last month, may no longer describe today's market.
Systems that update volatility every day, using Engle-style models, give a more honest picture of the danger the firm faces.
In practice
Real-world examples.
Example
A bank risk manager uses a GARCH model to update daily value at risk. After a sharp fall in the market, the model automatically raises the estimated risk and the trading limits tighten. When markets calm down, the limits relax again without anyone having to intervene.
Example
An options trader uses a volatility forecast built on Engle's ideas to decide whether an option's price is high or low relative to expected market swings. If the forecast is lower than the volatility implied by the option price, she may sell the option.
Example
A regulator reviewing systemic risk uses methods developed by Engle to estimate how much capital a large bank might need in a severe downturn. The results help decide whether the bank must hold extra reserves.
Formula
Calculation
ARCH(1) variance forecast: Variance today = a0 + a1 x (Yesterday's Shock) squared
Long-run variance = a0 / (1 - a1)
Worked example: Assume a0 = 0.0001 and a1 = 0.4. Yesterday the portfolio fell 3%, so the shock is 0.03.
Variance today = 0.0001 + 0.4 x (0.03 x 0.03) = 0.0001 + 0.4 x 0.0009 = 0.0001 + 0.00036 = 0.00046
Standard deviation = square root of 0.00046 = about 0.02145, or 2.145%
Long-run variance = 0.0001 / (1 - 0.4) = 0.0001 / 0.6 = 0.0001667, which gives a standard deviation of about 1.29%.
On a $1,000,000 portfolio, a typical daily move after a 3% shock is forecast at $21,450, compared with about $12,900 on an ordinary day. That shows how a big move raises the risk estimate.Case study
Seen in the real world.
Calder Funds is a fictional investment firm that used a fixed estimate of 1% daily volatility to set its risk limits. In this illustrative case, markets fell several days in a row and actual daily moves reached 3%, yet the limits stayed the same.
A risk analyst rebuilt the system using an ARCH-style model, where the risk estimate rose after each large move. In a back-test on the firm's own history, the new method predicted the turbulent weeks far better and would have cut positions earlier.
The chief risk officer adopted the model for daily monitoring while keeping the old estimate as a simple cross-check. The firm later noted fewer limit breaches and steadier results during volatile periods.
Watch out
Common mistakes.
- Believing the model predicts market direction. It forecasts the size of moves, not whether prices rise or fall.
- Assuming volatility is constant. Engle's central insight is that risk changes through time and clusters.
- Mixing up ARCH and GARCH. GARCH extends ARCH by adding the previous variance forecast, which usually makes the model more compact.
Questions
People also ask.
What did Engle win the Nobel Prize for?
He shared the 2003 prize for methods of analysing economic time series with time-varying volatility.
Who shared the prize with him?
Clive Granger, who was honoured for work on trends shared between time series.
Why does a finance professional need to know this?
Volatility forecasts underpin risk limits, option pricing and regulatory capital, so the models are used throughout the industry. Even if you never run the model, you will meet its results in risk reports.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%Related