BMAN23000A Chap.7 Measuring Risk and Return
Measuring Risk and Return
Measuring Risk and Return covers Risk and Return Part 1, the opening of the block that runs through the second half of Semester 1. The lecture starts from history: shares have delivered the highest long-run returns and the largest fluctuations, so the question is how much return investors require per unit of risk. It then teaches three approaches to measurement.
The probabilistic approach uses a distribution of possible returns: expected return is the probability-weighted average and variance the expected squared deviation, with the standard deviation called volatility. Because true distributions are unknown, the historical approach uses realised returns, dividend yield plus capital gain with dividends reinvested, compounded across sub-periods, then averaged.
The sample variance divides by T minus 1, and the standard error of the average, volatility over the square root of T, gives a 95% confidence interval of two standard errors either side, which shows how imprecise a single stock's history is. The risk-based approach separates firm-specific risk, which diversification removes, from systematic risk, which remains.
Only systematic risk earns a premium, so beta, the sensitivity of a security's excess return to the market's, becomes the relevant measure, and the expected return is the risk-free rate plus beta times the market risk premium, a first look at CAPM.
What this chapter covers
- 01
Probability distributions, expected return and variance
- 02
Volatility as the standard deviation of returns
- 03
Realised returns and compounding sub-period returns
- 04
Average return and the sample variance
- 05
Standard error and the 95% confidence interval
- 06
Firm-specific versus systematic risk
- 07
Beta and the market risk premium
Worked example · free
Compute a realised annual return from quarterly data
- 2Compound the quarters: (1.04)(0.97)(1.06)(1.02) − 1 = 9.07% for the year.
- 1Average quarterly return = (4 − 3 + 6 + 2)/4 = 2.25%.
- 2Four times the average, 9%, differs from the compounded 9.07% because compounding earns returns on returns; with volatile returns the arithmetic average can overstate realised growth.
Key terms
- Volatility
- The standard deviation of an asset's returns.
- Realised return
- The dividend received plus the price change over a period, divided by the starting price.
- Diversification
- Averaging out independent risks by holding many assets.
- Beta
- The sensitivity of a security's excess return to the excess return of the market portfolio.
Measuring Risk and Return FAQ
Why divide by T minus 1 for historical variance?
With realised data the average itself is estimated from the same observations, which uses up one degree of freedom. Dividing the sum of squared deviations by T minus 1 gives the sample variance the lecture and formula sheet use; dividing by T would understate it.
Why is a single stock's average return unreliable?
Its standard error, volatility divided by the square root of the number of observations, is large because individual stocks are volatile and histories are short. The resulting confidence interval for the expected return is very wide, so more observations are needed for precision.
Is volatility the right risk measure for one stock?
Not for a diversified investor. Much of a single stock's volatility is firm-specific and disappears in a portfolio, so investors are not paid for it. The premium depends on systematic risk, which beta measures.
Exam move
Work through the three approaches in order and keep the formulas apart. For the probabilistic approach, practise small tables of prices or returns with probabilities: convert prices to returns, compute the expected return, the squared deviations, the variance and the volatility, and say which number is volatility.
For the historical approach, compute realised returns including dividends, compound quarterly or monthly returns into an annual figure, and then compute the average, the sample variance with T minus 1, the standard error and the confidence interval for a short series. Repeat the interval for a longer series to see it narrow.
The announcements warn that slide formulas missing from the sheet must be memorised, and the standard error is one of them. For the risk-based approach, explain in two or three sentences why the index is less risky than its members and why diversifiable risk earns no premium, using the arbitrage argument. Practise computing beta from changes in stock and market returns, as in the workshop, and then a required return from beta.
The workshop spreadsheets also show that the first few stocks added to a portfolio cut volatility sharply while later ones barely move it; reproduce that pattern with the portfolio variance formula from Chapter 8. These skills feed the long risk-and-return question that appeared in both recent January papers.
Working through Measuring Risk and Return in BMAN23000A? Sia is AskSia’s AI Finance tutor — ask any BMAN23000A Measuring Risk and Return question and get a clear, step-by-step explanation grounded in how BMAN23000A is taught and assessed. Read this chapter free, then take your hardest questions to Sia.