AQIF study plan

Compute and interpret mean, median, variance, skewness, and kurtosis for a given return series both by hand and in code

Classify a stated financial problem as discrete or continuous and select the matching probability distribution with justification

Apply conditional probability and Bayes' theorem to update a prior belief with new evidence

Derive confidence intervals for a mean and a proportion, naming the sampling distribution used in each case

Test hypotheses on means and variances and explain the Type I versus Type II error trade-off in a stated decision context

Fit simple and multiple linear regression models and diagnose heteroskedasticity and multicollinearity

Explain the bias-variance trade-off and map it onto concrete underfitting and overfitting symptoms in a dataset

Clean a small return dataset while documenting every missing-value and outlier decision

Implement and compare at least three supervised models (linear, regularized, tree-based) using time-series-aware validation splits

Compute precision, recall, and ROC-AUC, then justify which metric fits the stated cost of errors

Apply L1 and L2 penalties and quantify the resulting change in feature weights

Argue when clustering or PCA adds value to factor and portfolio work and when it does not

Compute portfolio expected return, variance, and Sharpe ratio from weights and a covariance matrix

Construct an efficient frontier and locate the tangency portfolio given a risk-free rate

Decompose total risk into factor and idiosyncratic parts and interpret marginal contribution to risk

Estimate parametric and historical VaR and CVaR and critique the assumptions behind each

Quantify bond price sensitivity using duration, convexity, and DV01 for a hypothetical fixed-coupon bond

Distinguish forwards, futures, options, and swaps by payoff shape and cash-flow timing

Apply put-call parity and identify the arbitrage trade when it is violated

Explain how volatility, skew, and term structure shape option prices in intuitive terms

Design a backtest specifying point-in-time data, transaction costs, and universe definition to avoid look-ahead and survivorship bias

Diagnose overfitting using out-of-sample, walk-forward, and deflated Sharpe ratio reasoning

Run parameter sensitivity analysis and report whether performance is stable or fragile

Audit a hypothetical backtest for at least five common biases and document a specific fix for each

Apply an ethical decision framework to conflicts of interest, fair dealing, and market-manipulation scenarios

Identify disclosure, record-keeping, and suitability obligations and distinguish them from voluntary best practice

Design a model governance process covering documentation, independent validation, and monitoring triggers

Produce exam-format written answers under time pressure and self-score them against a rubric

Complete two full-length timed mock exams and tabulate accuracy by domain

Re-derive from memory the 30 highest-yield formulas spanning statistics, risk, and ML metrics

Condense the two weakest domains into a one-page decision tree each

Rehearse a fixed time-allocation strategy for multi-part quantitative questions