Iran Official Insurance Actuary Exam common mistakes
Iran Official Insurance Actuary Exam · Understand the traps and how to avoid them.
On probability questions, computing a marginal or conditional probability without first writing down the joint support and the conditioning event, then reporting a number that integrates over the wrong region.
Candidates memorize formulas for marginals and Bayes' rule but skip the step of sketching the sample space; joint densities with non-rectangular support (e.g., 0 < x < y < 1) are common in this exam and the integration limits are where the marks are lost.
What to do instead
Before any computation, write the support explicitly as inequalities and, for conditional probability, state the conditioning event and its probability as a separate line. For every joint density problem, draw the region and derive the limits from the drawing, then check that the density integrates to 1 over that region as a sanity check.
Treating the force of interest, the effective annual rate, and the nominal rate convertible m times as interchangeable, and discounting cash flows at the wrong rate or over the wrong time index.
The exam mixes accumulation functions, present value factors, and annuity-immediate/annuity-due conventions; candidates under time pressure grab the first rate given rather than converting to a consistent basis, and off-by-one timing errors in annuity-due versus annuity-immediate go unnoticed.
What to do instead
At the top of each financial mathematics problem, build a small conversion table: i, d, delta, i^(m), d^(m), v, and the accumulation factor, all expressed from the single rate given. State whether each cash flow is at the beginning or end of the period, and verify the final answer by re-discounting a single cash flow at a time.
Using a single life table and a single interest rate for joint-life, last-survivor, and contingent (reversionary) annuity questions without adjusting for the dependency structure, producing answers that ignore the second life or assume independence where the contract does not.
Candidates learn the standard whole life and term insurance formulas well but practice joint-life and contingent benefits far less; the notation for statuses like (xy) and (x|y) looks similar and the survival probability formulas differ in ways that are easy to conflate.
What to do instead
For every multi-life problem, first write the event whose probability is needed in words (both alive at t, at least one alive at t, x alive and y dead), then translate to products or differences of survival probabilities. Keep a one-page formula sheet for joint-life and last-survivor statuses and re-derive each from the event description rather than recalling it.
Applying credibility to a single insured's experience without checking the exposure and heterogeneity conditions, and mixing up the limited fluctuation (classical) and greatest accuracy (Bühlmann) credibility formulas and their assumptions.
Credibility is taught as a formula to plug into, so candidates do not internalize when each model applies; the exam deliberately presents small-sample and large-sample settings, and the two approaches give different weights that candidates swap under pressure.
What to do instead
Before computing, classify the setting: is the standard for full credibility based on frequency or severity, and is the model limited fluctuation or Bühlmann? Write the credibility factor Z for the chosen model, state the assumptions (homogeneous risk classes, known within-class variance), and check that Z lies in [0,1]. Practice deriving the Bühlmann factor from the between- and within-class variance components rather than memorizing it.
In ruin theory, computing the adjustment coefficient or ruin probability using the wrong premium loading or the wrong claim-size distribution, and confusing the surplus process S(t) = u + ct - claims with the aggregate loss process.
The classical risk process has several equivalent parameterizations and candidates lose track of whether c is the premium rate or the safety loading; Lundberg's inequality and the exact ruin probability for exponential claims are both tested and the conditions for each are glossed over.
What to do instead
Write the surplus process explicitly with every symbol defined before substituting numbers, and derive the adjustment coefficient R from the equation E[e^{-R(X - cT)}] = 1 (or the equivalent), then state which bound or exact formula applies. For exponential claim sizes, derive the exact ruin probability rather than recalling it, and verify that the loading factor is positive.
Answering insurance economics questions by intuition about 'fairness' or 'risk pooling' without naming the specific market-failure concept (adverse selection, moral hazard, principal-agent, or the law of large numbers) that the question is testing.
These questions are conceptual and candidates revert to general reasoning; the exam expects the precise term and its mechanism, and graders award marks for identifying the mechanism and its consequence, not for a plausible narrative.
What to do instead
For each scenario, identify the asymmetric-information structure first: who knows what, before or after contracting. Then name the concept and state the predicted equilibrium effect (e.g., adverse selection leads to unraveling of low-risk pools). Keep a one-page glossary linking each concept to a canonical example and to the pricing or contract design remedy.
Citing Iranian insurance regulation from memory with outdated or invented article numbers, or applying a general principle without reference to the specific governing body (Bimeh Markazi Iran) and the applicable law.
Regulatory content changes and candidates rely on old study notes; the exam tests whether the candidate knows which authority regulates what, and fabricated citations are heavily penalized by graders who check the legal basis.
What to do instead
Build a personal regulation index: for each topic (solvency margin, reinsurance cession, tariff approval, social security contributions), record the governing instrument and the responsible body, and verify against the current official text. In answers, state the authority and the instrument by name; if the exact article is uncertain, describe the rule accurately without inventing a number.
In accounting and ALM questions, mixing statutory (regulatory) and economic balance-sheet treatments, or reporting a solvency ratio computed on the wrong capital base.
Insurers report under regulatory accounting while actuaries also reason on an economic basis; candidates conflate the two, and the exam deliberately tests the difference between available and required capital, and between book and market values of assets backing reserves.
What to do instead
For every balance-sheet or solvency question, label the basis (statutory versus economic, book versus market) at the top and keep it consistent throughout. Show the numerator (available capital) and denominator (required capital or solvency margin) separately, and check the result against the regulatory minimum before writing the conclusion.
Ignoring the professional-ethics dimension of a scenario and answering purely as a technical calculation, or failing to identify the conflict of interest and the required disclosure and escalation path.
Candidates treat ethics as a soft topic and spend their time on computation; ethics questions in this exam are scenario-based and expect the candidate to name the standard, the duty owed, and the action (disclose, decline, escalate, document).
What to do instead
For each scenario, run a fixed checklist: who is the client, what is the conflict, which standard or code applies, what must be disclosed and to whom, and what must be documented. Write the answer in that order, naming the duty (competence, integrity, confidentiality, objectivity) explicitly rather than describing it in general terms.
Presenting a final numeric answer with no audit trail: no stated assumptions, no intermediate values, and no units, so that a single arithmetic slip or a defensible alternative assumption loses the whole question.
The exam is time-pressured and candidates optimize for finishing; they also assume the marker only checks the final number, whereas actuarial marking schemes award method marks for stated assumptions and intermediate steps.
What to do instead
Adopt a fixed answer template for every computational question: assumptions and data given, formula, substitution, intermediate result with units, final answer with a reasonableness check. Reserve the last two minutes of each question to confirm that every symbol used has been defined and that the answer's magnitude is plausible.