MANCOSA · PGDip
RM8 Risk Modelling — Model Answers
Complete Module closed — notes, packs and references all shipped.
Worked solutions for the summative, formative and project assessments, 2019 to 2026.
From the module
A page from the RM8 material
Set exactly as it appears in the document.
4.2Five measures of risk, computed on the same data
MANCOSA, Final Online Summative Assessment, Risk Modelling, 17 June 2020, Question 4.2. The paper supplies no data.
Required (15)
Discuss ANY FIVE (5) measures of risk in risk management.
Step 1 — The five measures
The paper supplies no data, so two portfolios of twenty equally likely annual loss outcomes in thousands of rand are declared and used. Portfolio A is symmetric; Portfolio B is right-skewed.
| Portfolio | The twenty outcomes (R'000) | Shape |
|---|---|---|
| A | 40, 42, 44, 46, 48, 49, 50, 50, 50, 50, 50, 50, 50, 50, 51, 52, 54, 56, 58, 60 | Symmetric, skewness +0.0000 |
| B | 44, 45, 46, 47, 47, 47, 48, 48, 48, 49, 49, 49, 50, 51, 51, 52, 53, 54, 56, 66 | Right-skewed, skewness +1.8769 |
| Measure | What it measures, and its limitation | Portfolio A | Portfolio B |
|---|---|---|---|
| Standard deviation | Total dispersion about the mean, in the units of the data. It counts favourable and adverse deviation equally, which for a loss variable is not what a manager wants | 4.7011 | 4.7011 |
| Semi-deviation | Dispersion on the adverse side only. It answers the standard deviation's main objection, and it is harder to interpret because it has no equivalent of the empirical rule | 3.3242 | 4.0187 |
| Range | The distance from best to worst outcome. The crudest measure, entirely determined by two observations and therefore unstable | 20 | 22 |
| Percentile, or value at risk | The loss not exceeded at a stated confidence — here the 95th percentile, so nineteen years in twenty. It is a threshold and says nothing about what lies beyond it | 58.1000 | 56.5000 |
| Expected tail loss | The average loss given that the percentile has been breached. It answers the percentile's objection and requires more data, because it is an average over the fewest observations | 60.0000 | 66.0000 |
Step 2 — What the computation shows
The two portfolios have exactly the same mean of 50 and exactly the same standard deviation of 4.7011, so the coefficient of variation is identical too at 9.40 per cent. On the three measures a candidate is most likely to name first, they are indistinguishable. On the other two they are not, and — this is the finding worth the marks — the measures do not merely differ in sensitivity, they disagree about which portfolio is riskier.
| Measure | Portfolio A | Portfolio B | Verdict |
|---|---|---|---|
| Standard deviation | 4.7011 | 4.7011 | Identical |
| Coefficient of variation | 9.40 % | 9.40 % | Identical |
| Semi-deviation | 3.3242 | 4.0187 | B is riskier |
| 95th percentile | 58.1000 | 56.5000 | A is riskier |
| Expected tail loss | 60.0000 | 66.0000 | B is riskier |
Three different answers from five measures on the same two portfolios. The percentile ranks A as the worse of the two because A's distribution is wider through its body, while the expected tail loss ranks B worse by 6.00, or 10.00 per cent, because B's worst outcome is 6 higher. That disagreement is not a defect in the data; it is the standard criticism of value at risk, demonstrated on twenty numbers — a percentile is a single point on a distribution and is silent about everything beyond it.
Step 3 — Choosing between them
The judgement the fifteen marks want is a rule. Use the standard deviation for a symmetric variable where both directions matter and where the measure must feed a further calculation, since almost every other quantitative technique takes it as an input. Use the semi-deviation where only the adverse side matters and the audience can interpret it. Use the range only as a sanity check. Use a percentile where a threshold must be set — a contingency, a limit, a capital number — and always state the confidence level with it. And use the expected tail loss wherever solvency rather than profitability is the question, because that is the only one of the five that describes what happens once the threshold has been passed.
'Two portfolios can share a mean of 50 and a standard deviation of 4.70 exactly and still differ by 10 per cent in expected tail loss — so the question is never which measure is correct but which question the decision is actually asking.'
How it is examined
How RM8 is examined
We are confirming the current assessment format for RM8. Tell us your sitting and we will send it with your free extract.
Module facts
| Module code | RM8 |
|---|---|
| Institution | MANCOSA |
| Level | PGDip |
| Prescribed text | Smith & Merritt, Proactive Risk Management (CRC, 2020); Rees, Business Risk and Simulation Modelling in Practice (Wiley, 2015) |
| Part of | MANCOSA Postgraduate Diploma in Risk Management |
| Updated | 28 September 2026 |
Questions
RM8: what students ask
Is there an exam pack for RM8 with worked solutions?
Yes. The complete RM8 module includes the past papers and assignments held for it, worked in full. Ask which sittings are held for your examination and we will tell you before you pay.
What does the RM8 material include?
Worked solutions for the summative, formative and project assessments, 2019 to 2026.
How much does RM8 cost?
R1 450 for the complete module, R850 for the notes only, or R320 for a single sitting or assignment pack. Tutoring from the same pages is R450 an hour.
Can I see a sample of the RM8 material before I pay?
Yes. A full page from the RM8 material is shown on this page, and you can ask for a free extract from the module. If the first twenty pages of what you order do not earn their price, the payment is returned in full.
Is the RM8 material up to date for 2026?
The RM8 page was last updated on 28 September 2026, and a complete-module order includes the updates issued for your sitting year. Tell us your sitting when you order and we confirm what covers it.
How do I pay for RM8, and how quickly do I get it?
Order on this site and an invoice arrives by email. Pay by EFT with the reference on it, and the material is emailed to you, normally within one working day of the payment reflecting.
Which qualification is RM8 part of?
RM8 Risk Modelling is a module of the MANCOSA Postgraduate Diploma in Risk Management. The programme page lists every module in it and what is available for each.
Is TheronGroup part of MANCOSA?
No. TheronGroup of Consultants is an independent tuition and study-material provider, not affiliated with or endorsed by MANCOSA. The module code identifies the subject you are studying; your tutorial letter or module guide is the authority on what is examined.
See the MANCOSA PGDip Risk Management programme page for every module in it.
From the blog
Written about RM8
Related
Other MANCOSA modules
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.
Free extract
A full worked example and a note extract from the module you are sitting, sent the same week you ask.
Employers and professional bodies
In-house governance, risk and internal-audit training, and bulk licensing for staff in part-time postgraduate study.