TheronGroup of Consultants

80/20 Golden Rule · Pareto Optimality

Worked example

RSK4805: futures pricing, swaptions, option Greeks, duration and EWMA

UNISA's May/June 2024 RSK4805 Question 2 worked in full: a mispriced future, swaptions, the Greeks on a short option book, bond duration and an EWMA update.

UNISA · RSK4805 Market Risk Management · May/June 2024 · Question 2

Question 2 of UNISA's May/June 2024 RSK4805 Market Risk Management paper carries 30 marks across six short parts, and between them they cover most of the calculations the module examines. Each is quick once the formula is secure, and each has one trap. The questions are described in our own words below; the answers are TheronGroup's.


2.1.1Is the future mispriced?6 marks

A company’s shares trade at R100. The continuously compounded risk-free rate is 9,5% a year, and a six-month futures contract on the share trades at R101. Calculate the price the future should trade at, and say whether it is overvalued, undervalued or correctly priced, with your reason.

The no-arbitrage futures price on a non-dividend-paying stock is the spot price compounded forward at the risk-free rate. The actual price is then compared with it.

F0 = S0 × e^(r × T)
   = R100 × e^(0,095 × 0,5)
   = R100 × e^0,0475
   = R100 × 1,048646
   = R104,8646

Actual futures price        = R101,00
No-arbitrage futures price  = R104,8646
Gap                         = R3,8646 below
  • The contract is undervalued: at R101,00 it trades R3,86 below the R104,86 that the cost of carry requires.
  • The futures price must equal the cost of buying the stock today and financing it to the delivery date. At any lower price, a trader can sell the stock short, invest the proceeds at 9,5% and buy the stock back through the future for less than the investment has grown to.
  • That trade locks in R3,86 a share with no net investment and no price risk, which is why the price cannot persist.

Where the marks are lost. Continuous compounding means e^(rT); using (1 + r)^T gives R104,64 and a different gap. Six months is T = 0,5, not 6. And say which way the arbitrage runs: an undervalued future means buy the future and sell the stock.

2.1.2Swaptions4 marks

Explain what a swaption is, and name the two types.

  • A swaption is an option on an interest rate swap. It gives the holder the right, but not the obligation, to enter into a swap at a specified fixed rate on or before a specified date, and like any option it is paid for with a premium up front.
  • It is used to fix the worst case on a borrowing or investment decision not yet made. A company expecting to borrow in six months can buy the right to swap into a fixed rate now and abandon it if rates have fallen by then, so the swaption caps the rate without committing to it.
  • A payer swaption gives the right to enter the swap as payer of the fixed rate and receiver of the floating rate. It is exercised when rates rise above the strike, so it protects a borrower.
  • A receiver swaption gives the right to enter the swap as receiver of the fixed rate and payer of the floating rate. It is exercised when rates fall below the strike, so it protects an investor or lender.

Where the marks are lost. Payer and receiver are named for the fixed leg, not the floating leg; getting that the wrong way round reverses both answers. Naming the two types alone earns only two of the four marks.

2.2.1The Greeks on a short option position6 marks

A trader is short 10 000 European call options. For one option: value R12,49, delta 0,600, gamma 0,015, vega 0,405 per percentage point, theta −0,045 per day, rho 0,373 per percentage point. For the short position of 10 000: value −R124 900, delta −6 000, gamma −150, vega −4 050, theta 450, rho −3 730. Explain the effect on the short position if (a) the share price rises by R0,20; (b) volatility rises by 0,70%; (c) one day passes with no other change. Show the calculations, and say whether the option price rises or falls and what that does to the position.

Each Greek is the change in the option’s value for a unit change in its own variable. The position is short, so its exposure has the opposite sign to the option’s, and the position figures are already given.

Scenario Greek Effect on one option Effect on the short position
a. Share price rises by R0,20 Delta 0,600 0,600 × R0,20 = R0,1200 increase −6 000 × R0,20 = −R1 200
b. Volatility rises by 0,70% Vega 0,405 0,405 × 0,70 = R0,2835 increase −4 050 × 0,70 = −R2 835
c. One day passes Theta −0,045 −0,045 × 1 = R0,0450 decrease 450 × 1 = R450
  • (a) The option price rises by R0,1200 when the share rises by R0,20; the trader is short, so the position loses R1 200.
  • (b) The option price rises by R0,2835 when volatility rises by 0,70%; the position loses R2 835.
  • (c) The option price falls by R0,0450 as a day passes; the short position gains R450, because time decay works for the writer.
  • The pattern is the whole point: a short option position loses on a favourable move in the underlying and on rising volatility, and earns its living from the passage of time.

Where the marks are lost. Theta is quoted per day and vega per percentage point; converting either is the standard error. The question asks for the effect on the option and on the position, and marks both.

2.2.2Duration and the bond price4 marks

A five-year bond with a continuously compounded yield to maturity of 11% pays an 8% annual coupon. Its price is R86,80 and its duration 4,75. What does a 0,3% rise in yield do to the price?

For a small parallel move, the change in a bond’s price is minus its duration, times the price, times the change in yield.

ΔB = −D × B × Δy
   = −4,75 × R86,80 × 0,003
   = −R1,2369

New price = R86,80 − R1,2369 = R85,5631

The price falls by R1,24, from R86,80 to about R85,56: a 0,3% rise in yield costs about 1,4% of the bond’s value, which is the duration of 4,75 times the 0,3% move.

Where the marks are lost. Express the yield change as 0,003, not 0,3, or the answer is a hundred times too large. The minus sign carries a mark: yields up, prices down. The coupon and maturity only make the duration plausible; they do not enter the calculation.

2.3.1Updating volatility with EWMA6 marks

An asset closed yesterday at R500 with an estimated volatility of 1,6% a day, and closed today at R495. Update the volatility estimate using the EWMA model with λ = 0,95, working to eight decimal places.

The exponentially weighted moving average carries yesterday’s variance forward at lambda and adds the latest squared return at one minus lambda.

Day's return:     un   = (R495 − R500) ÷ R500 = −0,01000000
                  un²  = 0,00010000

Yesterday:        σ(n−1) = 0,016   so σ(n−1)² = 0,00025600

Updated variance: σn² = λσ(n−1)² + (1 − λ)un²
                      = 0,95 × 0,00025600 + 0,05 × 0,00010000
                      = 0,00024320 + 0,00000500
                      = 0,00024820

Updated volatility: σn = √0,00024820 = 0,01575436 = 1,5754% a day

The updated estimate is 1,58% a day, down from 1,6%. The day’s move of −1% was smaller than the 1,6% volatility prevailing, so the estimate falls; with lambda at 0,95, only 5% of the weight sits on the new observation.

Where the marks are lost. Show every step: the return, the two weighted terms, the variance, the square root and the units. Carry eight decimals and round only at the end. EWMA has two terms; adding a constant term turns it into GARCH, which answers a different question.

2.3.2The correlation matrix4 marks

Define a correlation matrix, and explain why its diagonal elements always equal one.

  • A correlation matrix is a square, symmetric table of the correlation coefficient between every pair of variables in a set, so the entry in row i and column j is the correlation between variable i and variable j.
  • It is symmetric because the correlation between i and j equals that between j and i, and every entry lies between −1 and +1.
  • Each diagonal element is the correlation of a variable with itself. A variable moves exactly with itself, so the coefficient is one by construction.
  • It is the input the model-building approach needs: with the volatilities, it produces the variance-covariance matrix from which the portfolio’s standard deviation, and so its value at risk, is computed.

Where the marks are lost. Four marks means four points: the definition, the symmetry, the diagonal and the use. A correlation matrix must also be positive-semidefinite; where a question raises that, it is asking whether the correlations stated could all hold at once.

Studying RSK4805

Every part of this question is a standard formula applied with one trap in the units. The RSK4805 course has seven parts of notes, a formula and technique sheet and 18 solution packs covering the sittings and online assessments, from R200 a month.

Get in touch

Tell us your module

Send the module code and your sitting. We reply with exactly what exists for it today, a free extract from that module, and what it costs.

Students

Name your module code and sitting. You get a free extract from your own module — not a generic brochure — before any payment.

Send an enquiry

Free extract

A full worked example and a note extract from the module you are sitting, sent the same week you ask.

Request the extract

Employers and professional bodies

In-house governance, risk and internal-audit training, and bulk licensing for staff in part-time postgraduate study.

Corporate enquiry