Objectives of backtesting an investment strategy

Objectives of backtesting an investment strategy

Objectives of backtesting an investment strategy

On this page

Learning Outcome Statement

Describe objectives in backtesting an investment strategy.

On the exam, you will get a client type, portfolio objective, benchmark, constraints, and a short backtest description. Then you will choose the appropriate objective or portfolio conclusion in an MCQ or item set.

A backtest asks a practical question. If the client had used the strategy in the stated setting, could the strategy have helped the client pursue the portfolio objective? A backtest does not certify that the future will copy the past. You use the evidence to decide whether the strategy merits more investment work and whether the client can implement it within the stated constraints.

Decision lensBacktest questionPortfolio consequence
ObjectiveCould the strategy advance the client’s portfolio objective?Advance to further review only if its role supports the objective.
ConstraintCan the client implement the strategy?Reject an infeasible strategy even when its backtest looks desirable.
BenchmarkWhat makes the comparison relevant?Judge the evidence against the stated benchmark.
Risk willingnessWhat risk does the client choose to accept?Record governance preference, not financial ability.
Risk capacityWhat loss can the client financially absorb?Keep exposure within the client’s liquidity and funding ability.
Active riskHow far may the strategy depart from the benchmark?Assess benchmark-relative deviation separately.
Total portfolio riskWhat loss can the whole portfolio bear?Match total exposure to obligations and resources.

Objective and constraint answer different questions. A strong backtest can make a strategy look desirable, but the client’s inability to implement it gives the trap away. Active risk describes departure from the benchmark, while total portfolio risk describes the whole portfolio’s loss exposure. The benchmark makes the comparison relevant, and risk willingness captures preference while risk capacity captures financial ability.

Treat the formula as a screening rule rather than a forecast equation. It gives you a gate for deciding whether a strategy deserves further review. The strategy moves forward only when it fits the objective and can be implemented.

$$ \text{Further review} = \text{objective fit} \quad \text{and} \quad \text{implementation feasibility} $$

Further review is not approval to allocate. Objective fit supports more work when the strategy could advance the client’s objective. Implementation feasibility supports more work only when liquidity, funding capacity, and other stated constraints support the strategy. If either term fails, stop before further due diligence.

When you turn the backtest into a portfolio conclusion, walk through these checks. Start with the client’s objective, use the benchmark to make the comparison relevant, and test the stated constraints. If either gate fails, stop before further due diligence.

Loading diagram…

Worked example

You are advising an underfunded defined-benefit pension. The pension committee is considering a quality-equity strategy and wants a backtest against a broad equity benchmark before commissioning more work. Benefit-payment liquidity and funding capacity constrain what the portfolio can carry. The exam asks what the backtest should establish.

GivenCase fact
InvestorUnderfunded defined-benefit pension
ObjectiveSupport benefit payments and funding progress
StrategyQuality-equity strategy
BenchmarkBroad equity benchmark
ConstraintBenefit-payment liquidity and funding capacity
DecisionWhether further due diligence is warranted

The case facts identify the decision before you judge the evidence: this pension is testing a quality-equity strategy for a defined role, against a relevant equity benchmark, while meeting benefit payments. A favorable comparison matters only if the strategy supports funding progress and the pension can carry it within its liquidity and funding limits. Use each fact to answer one of those two gates.

  1. Start with the client objective by asking whether quality equity could advance benefit payments and funding progress. This connects the backtest to the pension’s reason for investing. The result produced is an objective-fit conclusion.

  2. Compare the quality-equity strategy with the broad equity benchmark. This shows whether the strategy adds value relative to the equity role it is meant to fill. The result produced is benchmark-relative evidence.

  3. Test feasibility by asking whether benefit-payment liquidity and funding capacity allow the pension to implement the strategy. This separates practical investability from an attractive objective fit. The result produced is a constraint conclusion.

  4. Combine the objective-fit and feasibility conclusions. This identifies whether the evidence supports more investigation, while avoiding an automatic allocation or a claim about future success. The result produced is a due-diligence decision.

The appropriate portfolio conclusion is that quality equity may deserve further due diligence if it fits the objective and constraints, not that it should be allocated automatically.

Changed condition

Keep the pension, objective, strategy, benchmark, liquidity, and funding capacity unchanged. Change only the committee’s willingness to tolerate active risk from cautious to high. Predict the portfolio conclusion before reading the table.

Decision pointBeforeAfter
Active-risk willingnessCautious committee preferenceHigh committee preference
Total-risk capacityLimited by underfundingLimited by underfunding
Portfolio conclusionKeep deviations tightly reviewedActive-risk review may widen, capacity unchanged

The prediction should be that the active-risk assessment moves because governance willingness changed. Total portfolio risk does not move because underfunding, benefit-payment liquidity, and funding capacity did not change. A willingness to depart from the benchmark cannot create the ability to absorb total portfolio losses.

Before reading on, write the active-risk conclusion and the total-risk conclusion separately.

Common error

You may see the committee’s high willingness to tolerate active risk and recommend a portfolio with high total risk. That choice feels right because the committee judges the strategy against a benchmark and has approved active deviation. But the underfunded status, benefit-payment liquidity, and funding capacity show that the committee’s willingness is not its capacity, so you must assess active risk separately from total portfolio risk.

Check yourself

Practice

A university endowment has a spending objective. The committee accepts active risk, but near-term spending needs make liquidity important and the endowment has limited capacity to absorb total losses. An equity strategy is tested against a broad equity benchmark. The backtest is favorable, and the committee asks what conclusion it supports.

Question 1

Which objective is the backtest primarily addressing?

Question 2

Why is the broad equity benchmark included?

Question 3

Which statement best separates the committee's risk views?

Question 4

What is the most appropriate portfolio conclusion from a favorable backtest?

An underfunded defined-benefit pension tests a quality-equity strategy against a broad equity benchmark before deciding whether further due diligence is warranted. The committee is willing to tolerate active risk, but benefit-payment liquidity is tight. Which conclusion is most appropriate?

Revision snapshot

Backtesting informs an investment decision. It does not prove future success.

$$ \text{Further review} = \text{objective fit} \quad \text{and} \quad \text{implementation feasibility} $$

LensExam move
ObjectiveAssess whether strategy could advance client objective.
ConstraintJudge implementation feasibility separately from desirability.
BenchmarkUse relevant basis for evaluating strategy.
RiskSeparate willingness from capacity and active risk from total risk.
Exam angleChoose the conclusion fitting client constraints.

Risk willingness is not risk capacity. An underfunded defined-benefit pension tests a quality-equity strategy against a broad equity benchmark before deciding whether further due diligence is warranted.

Unlock the full lesson and continue reading

Signup to continue reading this lesson and unlimited access questions, flashcards, AI notes, and more

Scan to download app

Scan to download
UNLOCK FREE ACCESS

Practice Questions: Objectives of backtesting an investment strategy

  1. 1. The Meridian Pension Plan is reviewing the risk of its strategic portfolio. The plan's investment committee has instructed chief investment officer Elena Ruiz to distinguish capital allocation from risk allocation. The committee is particularly concerned that the largest asset weight may contribute a disproportionately large share of total portfolio variance. Ruiz uses the standard variance–covariance framework and assumes that the volatilities, correlation, and betas in Exhibit 1 are annual estimates. The portfolio is fully invested, with no derivatives or leverage. For purposes of calculating portfolio beta, each asset-class beta is measured relative to the same global equity market factor, and portfolio beta can be calculated as the weighted average of the asset-class betas. Ruiz reminds the committee that an asset's component contribution to portfolio variance equals its portfolio weight multiplied by its covariance with the portfolio. The component contributions sum to total portfolio variance. She cautions that using capital weights alone as risk contributions would ignore differences in volatility and covariance. The plan is also evaluating a proposed active equity mandate using a two-factor model. The model's factor exposures, factor variances, covariance, and portfolio-specific variance are presented in Exhibit 2. The model has been estimated using returns expressed in decimal form. Ruiz defines systematic variance as the variance explained by the two factors, including their covariance term. Total variance equals systematic variance plus portfolio-specific variance. One committee member argues that portfolio-specific variance should be included when reporting the mandate's systematic standard deviation because it remains part of the mandate's observed risk. Ruiz responds that specific variance must be included when measuring total risk but excluded when isolating risk attributable to common factors. Another committee member proposes changing the negative factor covariance to its absolute value, reasoning that all covariance terms represent risk. Ruiz notes that a negative covariance reduces aggregate factor variance and must retain its sign.

    • 2. The investment committee of the Meridian Balanced Fund is reviewing its strategic asset allocation and an externally managed active equity mandate. The fund's objective is to earn a long-term return exceeding inflation by 3.5% while maintaining moderate total risk. There are no leverage or short-selling positions in the strategic portfolio. Returns and risk estimates in Exhibit 1 are annual, and the committee assumes that the stated expected returns, volatilities, and correlation are appropriate forward-looking inputs. The chief investment officer asks analyst Helen Wu to calculate the strategic portfolio's expected return and standard deviation. Wu notes that the weighted average of the two asset-class volatilities is 12.2%, but she is uncertain whether that value appropriately reflects the portfolio's total risk. She also wants to determine how much of the portfolio's standard deviation is contributed by each asset class, and how the correlation assumption drives the size of the diversification benefit. For risk-contribution calculations, the committee uses component risk in standard-deviation units: the asset weight multiplied by the asset's covariance with the portfolio, divided by portfolio standard deviation. The committee also reviews an active global equity manager benchmarked against the MSCI World Index. Exhibit 2 presents the manager's forward-looking active exposures and expected implementation costs. The risk model contains a market factor, a value factor, and security-specific residual risk. The two factor returns are assumed to be uncorrelated, and residual returns are uncorrelated with both factors. All factor volatilities and residual risk are annualized standard deviations. The manager argues that the mandate should be assessed using the gross information ratio because management fees are negotiated separately from the investment process. Wu disagrees, stating that the committee should focus on the information ratio available to the fund after both management fees and expected transaction costs. The committee accepts Wu's position because its manager-selection policy defines implementable active return as gross active return less recurring management fees and expected transaction costs. No question requires the expected-return calculation from another question. All calculations should use the inputs supplied directly in the relevant exhibit, and intermediate results should be retained without rounding.

      • 3. Marc and Elise Laurent engage an adviser to revise their taxable investment portfolio. They have a 15-year primary horizon, moderate willingness to accept risk, and substantial employment income. Within two years, however, they expect to purchase a second residence and must retain a meaningful allocation to cash and short-term bonds. They also want portfolio returns sufficient to preserve purchasing power after taxes and planned withdrawals. The adviser translates these circumstances into the constraints shown in Exhibit 2. The portfolio includes two individual stocks purchased for USD 400,000 each. One is now worth USD 650,000 and has favorable long-term prospects. The other is worth USD 250,000, and the adviser's research indicates that its competitive position has deteriorated. Marc insists that the loss-position stock should not be sold until it returns to its purchase price, but he is willing to sell the appreciated stock immediately because doing so would lock in a successful investment result. He does not cite any research contradicting the adviser's outlook. The Laurents also intend to make a USD 150,000 charitable gift this year. The charity is qualified to accept publicly traded securities directly. The family's appreciated individual stock is highly concentrated relative to the rest of the portfolio. The adviser is considering whether to transfer securities in kind or sell securities and contribute cash. The Laurents are subject to capital gains tax, and no restriction prevents them from donating the appreciated shares. To implement the strategic allocation, the adviser expects to purchase a regional-equity exchange-traded fund. The ETF is thinly traded, although its underlying securities are substantially more liquid. During the first several minutes after the market opens, the ETF commonly displays wider bid–ask spreads and greater uncertainty concerning its intraday net asset value. The Laurents do not require immediate execution and prefer protection against purchasing at an unusually large premium. The adviser can use either market or limit orders. Finally, the adviser evaluates the three candidate portfolios in Exhibit 1. Expected returns and volatilities are nominal, annualized estimates over a full market cycle. Illiquid assets include private equity and directly owned real estate that cannot reliably fund the planned residence purchase. The policy constraints are binding, and the adviser must select a portfolio satisfying every stated constraint rather than merely choosing the portfolio with the highest expected return.

        • 4. The Republic of Arandia exports oil and maintains a sovereign wealth fund with two separately governed portfolios. The Stabilization Portfolio supports the government budget when realized oil revenue falls below the amount assumed in the fiscal plan. The Future Generations Portfolio invests surplus resource revenue for citizens after current oil reserves are depleted. Exhibit 1 summarizes their characteristics. The finance ministry expects a possible global slowdown during the next year. Under the ministry's downside scenario, oil prices and tax revenue would decline simultaneously, increasing withdrawals from the Stabilization Portfolio. These withdrawals would be made in Arandia's domestic currency. The fund's board therefore requires the stabilization assets to maintain a substantial domestic currency exposure and to avoid excessive transaction costs during stressed markets. The Future Generations Portfolio has no scheduled distributions for at least 15 years and is intended to maximize long-run purchasing power subject to its risk limit. The government cannot redirect this portfolio to ordinary annual budget spending without legislative approval. Its governing board accepts substantial short-term volatility and permits a larger allocation to illiquid investments because the portfolio has a long horizon and no immediate liquidity requirement. The chief investment officer is reviewing the assumptions in Exhibit 2. She notes that expected return alone is insufficient when selecting an alternative investment. Lockup period, cash-flow pattern, valuation uncertainty, diversification potential, and the portfolio's capacity to meet capital calls must also be considered. Core infrastructure is expected to make regular cash distributions after completion of an initial investment period. Venture capital is expected to make no recurring distributions during the early years and may require follow-on capital before any realizations occur. One board member proposes assigning the same strategic allocation to both portfolios to simplify governance. Another argues that the portfolios have fundamentally different liabilities and risk capacities. The risk team also warns that oil-related fiscal needs are countercyclical: the government's need for withdrawals is likely to be greatest when global risky assets are declining and market liquidity is impaired. The team recommends evaluating liquidity under stressed conditions rather than relying only on normal-market trading volumes. All expected returns and volatilities in Exhibit 2 are annual nominal estimates in USD, and taxes do not apply to either portfolio.

          • 5. The Arbordale Stabilization Fund is a sovereign wealth fund financed by past commodity revenues. Its mandate is to support annual government expenditures while preserving the real value of assets for future generations. The fund has no contractual pension liabilities, but annual transfers to the government are significant and predictable. The investment policy requires at least three years of forecast transfers to be held in high-quality government bonds and cash. That reserve is currently fully funded and is included in the fixed-income allocation shown in Exhibit 1. The governing board sets strategic targets and permitted ranges. For liquid public asset classes, a breach of a permitted range triggers rebalancing to the strategic target rather than merely to the nearest boundary. Private investments are valued quarterly and are not mechanically rebalanced. The policy instead prohibits new commitments that would cause total illiquid private investments to exceed 25% of fund assets. Existing unfunded commitments have already been incorporated into the reported 18% exposure, so no additional adjustment is required when calculating remaining policy capacity. A recent rise in global equity markets increased public equity above its permitted range. The investment staff can transact without materially affecting market prices and can use the proceeds to restore underweight liquid asset classes. The board is also considering a new private infrastructure commitment equal to 10% of fund assets. The proposed vehicle has a 12-year legal life, provides no contractual redemption right, and is classified as an illiquid private investment under the policy. The fund's domestic currency is not the euro. Its EUR-denominated fixed-income exposure is shown in Exhibit 2. Currency risk for developed-market foreign fixed income must be hedged within the stated range using forward contracts. The committee has selected an 80% hedge ratio, expressed as a percentage of the foreign-currency market value, and intends to sell euros forward against the domestic currency. Finally, the board reviews whether its spending rule remains consistent with preserving real capital. For this review, the board uses the additive approximation that sustainable spending equals the expected nominal portfolio return minus expected domestic inflation and annual investment-management costs. Taxes are not applicable to the sovereign fund.

            • 6. After three years of strong equity returns, an adviser raises the strategic equity allocation solely because recent performance is expected to continue. This judgment is most likely affected by:

              • A. Recency bias
              • B. Representativeness bias
              • C. Status quo bias
              Show answer

              Answer: A. Recency bias

              Recency bias - The adviser expects three years of strong equity returns to continue, giving recent performance too much weight when forecasting future returns. - Raising the strategic equity allocation solely on that expectation makes the recent experience, rather than a long-term assessment, decisive. Representativeness bias - Representativeness bias would involve judging future returns by how closely the market resembles a perceived pattern or category, potentially neglecting base rates. - Here, the stated reason is simply that recent performance is expected to continue, which points to recency bias. Status quo bias - Status quo bias favors keeping an existing decision or allocation unchanged. - The adviser raises the strategic equity allocation, so the judgment does not reflect a preference for the current allocation.

            Flashcards: Objectives of backtesting an investment strategy

            1. 1. What are the steps in backtesting?

              Show answer

              1. strategy design 2. historical investment simulation 3. analysis of backtesting output

            2. 2. What is growth backtesting based on?

              Show answer

              - consensus earnigns growth for the coming year - stocks with high earnings growth preferred

            3. 3. What are the metrics in backtesting?

              Show answer

              - sharpe ratio - sortino ratio - VaR - conditional Var - maximum drawdown

            4. 4. Monte Carlo simulation and the historical simulation approach both account for what?

              Show answer

              skewness, excess kurtosis and tail dependence

            5. 5. What is defensive value backtesting based on?

              Show answer

              - trailing earnings yield - high E/P stocks preferred

            6. 6. What is profitability backtesting based on?

              Show answer

              - ROE - high ROE preferred

            7. 7. What is price momentum backtesting based on?

              Show answer

              - based on 12-month total return, excluding the previous month - stocks with momentum are preferred

            8. 8. What is rolling window backtesting?

              Show answer

              - proxy for actual investing - use a walk forward system rather than dividing the data into just two samples - parameters are tuned over time but the primary model method is fixed (to avoid overfitting) - strategy accepted if it performs well in the out-of-test periods (and makes logical sense) - Shortcomings - assumes past performance repeats in the future - may not reflect nature of financial markets & exteme downside risk

            9. 9. What is leverage backtesting based on?

              Show answer

              - debt/equity ratio - low leverage stocks preferred

            Everything you need for CFA prep

            Get full Oncourse access with lessons, practice questions, flashcards and AI study tools.

            GET STARTED FOR FREE
            75% OFFLimited time offer
            GET 75% OFF