What Is Portfolio Optimization? Simple Explanation and Example

What Is Portfolio Optimization? (Simple Explanation + Example)

Portfolio optimization is the process of determining the best possible weighting of different assets (equities, bonds, gold, currency, funds, etc.) in an investor's portfolio according to a specific objective. The goal is to achieve higher expected return at the same risk level, or to reduce risk at the same expected return level. In short, it is the process of directing every unit of your capital to wherever it will produce the best overall outcome.

In finance literature, this approach is grounded in Modern Portfolio Theory. This theory states that not only the expected returns of assets but also their risks (volatility) and how they move relative to each other (correlation) must be evaluated together. A well-optimized portfolio, even if composed of individually high-risk assets, can deliver more stable and predictable performance when combined.

Simple Explanation: What Suits Your Savings Best?

Think of it in everyday terms: you can make the same wardrobe far more functional by combining different outfits. Portfolio optimization is similar. By bringing together different "pieces" like equities, bonds, and gold, you create a "combination" that adapts to various conditions and protects you from unnecessary risks. Committing your entire budget to a single product (say, only equities) is like assuming the weather will always be the same. But economic trends, interest rates, inflation, and market psychology are constantly changing. A well-optimized portfolio means being prepared for different weather conditions.

Core Concepts

Expected Return

Expected return is the projected average gain an asset is anticipated to deliver, based on historical data, analyst expectations, current interest rates, and the macroeconomic outlook. The portfolio's expected return is calculated by multiplying each asset's expected return by its portfolio weight and summing the results. For example, a portfolio of 50 percent equities, 30 percent bonds, and 20 percent gold has an expected return that is the weighted average of these assets' individual expected returns at those proportions.

Risk (Volatility) and Downside Risk

Risk is most commonly measured through the fluctuation of returns (volatility). High volatility means returns can vary over a wide range. In practice, however, investors do not mind "upside" surprises; the real concern is "downside" losses. For this reason, some measures specifically address downside risk (e.g., Sortino ratio ratio). Nonetheless, the most common starting metric in portfolio optimization is standard deviation (volatility) and the variance-covariance approach built on it.

Correlation and Diversification

Correlation describes whether two assets move in the same or opposite direction and to what extent they move together. When correlation is low or negative, one asset may rise while the other falls. This reduces total portfolio volatility. While diversification does not magically increase returns, it lowers risk for the same return target or makes higher returns possible at the same risk. Portfolio optimization aims to quantitatively identify where diversification delivers the most benefit.

Risk Tolerance and Investment Horizon

Every investor has a different pain threshold. Some can endure short-term volatility, while others want a calmer portfolio to sleep well at night. Risk tolerance is a primary input parameter for your asset allocation. When combined with the investment horizon (short, medium, long term), portfolio optimization yields more meaningful results. Those close to a short-term goal (like a down payment) select a lower-risk mix, while long-term goals (like retirement) may favor more volatile but higher-potential-return solutions.

Modern Portfolio Theory and the Efficient Frontier

Modern Portfolio Theory calculates a portfolio's expected return and risk (variance) from the expected returns, volatilities, and covariances of the assets. In this framework, each possible weight set maps to a return-risk pair. Plotting these points on a graph, the outermost curve of the "best" combinations is called the efficient frontier. Portfolios below the efficient frontier are not rational because a better alternative exists at the same risk.

Two special portfolios stand out on the efficient frontier:

• Minimum variance portfolio: The portfolio with the lowest volatility across the entire efficient frontier.
• Maximum Sharpe portfolio: The portfolio that best rewards increasing risk relative to return (measured against the risk-free rate).

The investor's selection point is determined somewhere on the efficient frontier based on their risk tolerance. A more risk-seeking investor moves toward the "higher return, higher risk" region, while a risk-averse investor gravitates toward the "lower return but calmer" side.

Metrics: Sharpe, Sortino, Beta, VaR, and CVaR

The Sharpe ratio measures how much return is earned per unit of risk by dividing the portfolio's excess return over the risk-free rate by its volatility. The Sortino ratio ratio applies the same logic but uses only downside volatility as the denominator. Beta shows how sensitive the portfolio is to market movements. VaR (Value-at-Risk) and CVaR (Conditional VaR) assess the maximum potential loss and its expected magnitude at a given confidence level. These metrics become especially meaningful when your portfolio contains leverage, derivatives, or concentrated individual risks.

Simple Example: Portfolio Optimization with 3 Assets

Let's make the concepts concrete. Suppose we are using three asset classes:

• Equity index fund (e.g., a BIST 30-weighted ETF/fund)
• Government bond fund (short-to-intermediate maturity)
• Gold fund (TRY-denominated)

Let's work with a simple, hypothetical dataset:

• Expected returns (annual): Equities 12%, Bonds 6%, Gold 10%
• Volatility (annual standard deviation): Equities 22%, Bonds 7%, Gold 15%
• Correlations: Equity-Bond 0.20; Equity-Gold -0.10; Bond-Gold 0.00

The portfolio's expected return is the sum of weights multiplied by these returns. Risk depends not only on individual risks but also on the correlations between them. For example, consider these two portfolios:

• Portfolio A: Equities 50%, Bonds 30%, Gold 20%. Expected return = 0.5x12% + 0.3x6% + 0.2x10% = 9.8%. With correlation effects included, calculated volatility is approximately 11.7%.
• Portfolio B (lower risk focus): Equities 25%, Bonds 55%, Gold 20%. Expected return approximately 8.3%, volatility approximately 7.7%.

In this example, Portfolio B offers lower return than Portfolio A but significantly reduces risk. On the other hand, if we invested solely in equities, expected return would be 12% but volatility would be 22%. Portfolios A and B substantially reduce volatility through diversification. Looking at a rough risk-independent performance measure like "return/volatility," Portfolio B can be seen as more efficient.

Thinking in terms of a target return approach: say you target at least 10% annually. Setting weights to equities 50%, bonds 25%, gold 25% produces an expected return of approximately 10%. At this composition, volatility is still around 11.7%. As you can see, by adjusting weights within the correlation structure, it is possible to create lower or higher risk profiles around the same return.

Important note: The figures here are entirely for illustrative purposes. In the real world, expectations change over time; factors like inflation, policy rates, and currency volatility can quickly update parameters in Turkish markets. This is why regular data updates and periodic rebalancing are critically important.

How to Apply It: Optimize Your Portfolio in 7 Steps

• Clarify your target: Set an annual target return and maximum tolerable drawdown. Define your time horizon (e.g., 3, 5, 10 years).
• Gather data: Calculate returns from monthly or daily price series for your selected funds/ETFs/assets. TEFAS funds, Borsa Istanbul ETFs, and gold/currency data are suitable starting points.
• Estimate parameters: Calculate expected return, volatility, and pairwise correlations for each asset. A 3-5 year lookback is generally a good starting point; however, shorter windows may be more relevant during regime changes.
• Choose a model: For a simple start, set up a mean-variance optimization scenario with minimum variance or target-return risk minimization. For more advanced work, you can try risk parity, Black-Litterman, or factor-based models.
• Define constraints: Add real-world limitations to the model, such as minimum-maximum weight per asset, cash allocation, liquidity constraints, transaction cost limits, and ethical/ESG filters.
• Backtest and stress test: Test portfolio behavior across past periods and different market conditions (high inflation, currency shock, rate hikes, crisis periods). Monitor Sharpe/Sortino, maximum drawdown, and recovery times.
• Implement and rebalance: Reach target weights, then rebalance at your chosen intervals (e.g., quarterly, semi-annually) or on a threshold basis (e.g., when an asset deviates more than 5% from target).

Different Optimization Approaches

Maximum Sharpe

Finds the portfolio with the highest risk-adjusted return above the risk-free rate. Suitable for investors who can tolerate high volatility if the return is satisfactory.

Minimum Variance

Targets minimizing risk rather than expected return. More suitable during periods of high market uncertainty or for investors with low risk appetite.

Allocates not by nominal weight but so that risk contribution is equally distributed. For example, lower-volatility bonds may receive higher nominal weight because their risk contributions are balanced against equities. This approach aims to reduce excessive dependence on a single risk factor.

Black-Litterman

Combines market equilibrium (CAPM) expectations with the investor's subjective views. For example, you can feed your positive/negative views on a particular sector or asset class into the model with an uncertainty parameter and achieve more balanced and stable weights.

Factor-Based Optimization

Decomposes returns into factors such as market, value, growth, quality, momentum, and low volatility. The goal is to achieve a more stable risk-return profile through diversification across factors.

Real-World Constraints and Turkey-Specific Notes

• Transaction costs and taxes: Frequent rebalancing increases commission and spread costs. On the tax side, details like withholding, fund type, and holding period affect net returns.
• Liquidity and tracking error: Some funds/ETFs may have limited liquidity. Tracking error in index-tracking products can affect performance.
• Currency and inflation dynamics: Currency and inflation volatility is high in Turkey. Using foreign currency, gold, and TRY interest products together can help preserve purchasing power.
• Asset universe: A wide universe exists including Borsa Istanbul ETFs, TEFAS funds, deposits, government bonds, eurobonds, and gold/FX deposits. Each has a different risk profile and tax treatment.
• Legal and operational constraints: Retirement funds (BES) have different fund sets and distribution rules. Brokerage/bank accounts offer more flexibility.

Common Mistakes and How to Avoid Them

• Optimizing only for past returns: Past best performance may not repeat. Keep parameters current and test across different scenarios.
• Assuming correlations are constant: Correlations can rise during market stress. Include stress periods in your dataset; use conservative assumptions.
• Excessive complexity: Adding too many assets and constraints can destabilize the model. Start simple; increase complexity gradually as needed.
• Neglecting rebalancing: The portfolio drifts from target allocation over time. Disciplined periodic or threshold-based rebalancing keeps risk under control.
• Dependence on a single factor: Portfolios carrying only equity risk or only interest rate risk struggle during regime changes. Build balance across factors.

Personalizing the Strategy: By Age, Income, and Goal

There is no single "best" portfolio. The optimal allocation differs for a 25-year-old with a long horizon and regular income versus a 55-year-old approaching retirement who prioritizes capital preservation. Younger investors may target long-term growth with higher equity weights, while those nearing retirement may relatively increase bonds, deposits, and gold. However, in a country with high inflation, focusing on real returns rather than nominal returns is essential. A real return perspective helps preserve purchasing power without taking excessive risk.

Rebalancing and Schedule

Rebalancing ensures the portfolio regularly returns to its target weights. Two basic approaches:

• Time-based: Automatic rebalancing every 3-6-12 months. Simple and disciplined.
• Threshold-based: Rebalancing when an asset deviates from its target weight by a certain percentage (e.g., ±5%). Can reduce transaction costs and unnecessary trades.

Rebalancing automates the discipline of "selling high and buying low." Overvalued assets' weights are reduced, underperformers' weights are increased, keeping the portfolio's risk profile constant.

Scenario Examples: Different Compositions for Different Goals

• Growth-oriented (high risk tolerance): Equities 60-70%, Bonds 10-20%, Gold/FX 10-20%. Goal is long-term capital growth; tolerance for short-term volatility is required.
• Balanced (moderate risk): Equities 40-50%, Bonds 30-40%, Gold/FX 10-20%. Seeks a balance between growth and protection.
• Income and preservation (low risk): Equities 10-25%, Bonds/Deposits 50-70%, Gold/FX 10-20%. Capital preservation and regular cash flow are prioritized.

These ranges are representative and should be adjusted based on market conditions, interest/inflation levels, and personal goals. Especially during periods of high interest rates, the role of fixed-income instruments in the portfolio may increase; however, attention to real returns after inflation and taxes is essential.

Practical Tools and Tips

• For data: Brokerage/bank apps, fund platforms, historical price series from index providers.
• For calculations: Expected return, volatility, and correlation matrices can be produced simply with spreadsheets (Excel/Google Sheets). Use Solver to find minimum variance or minimum risk for a target return.
• For robustness: Test parameters with different windows (1-3-5 years) and stress periods. Don't over-rely on a single period.
• For constraints: Set floor/ceiling weights for each asset (e.g., gold max 25%) to prevent the model from going to extreme weights.
• For costs: Preferring low expense ratio ETFs/funds improves long-term net returns.

Brief Look at Advanced Topics

More Stable Portfolios with Black-Litterman

Combines market equilibrium returns with your own views. For example, you can feed a view like "gold will remain relatively strong in TRY terms over the next 12 months" into the model with an uncertainty parameter, achieving a balanced adjustment rather than extreme weights.

Factor Allocation

Going beyond sector or country-based allocation on the equity side to allocate across factors like value, quality, growth, momentum, and low volatility reduces dependence on a single narrative.

CVaR Optimization

Focuses on unexpected but painful "tail risks." Especially meaningful for portfolios using leverage, containing derivatives, or with asymmetric profiles.

Frequently Asked Questions

Is portfolio optimization a one-time exercise?

No. Market conditions, interest rates, inflation, and correlations change. Parameters need to be updated and regular rebalancing is required.

How often should I rebalance?

For most investors, every 3-6 months or a ±5-10% threshold-based approach is reasonable. Adjust based on transaction costs and tax effects.

Which data window should I use?

Generally 3-5 years is a good starting point. However, shorter windows may better reflect current dynamics during major regime changes.

How do I estimate expected returns?

Historical averages, current interest rate and risk premium approaches, analyst consensus, and scenario analysis are common methods. Using reasonable ranges rather than relying on a single method is the healthiest approach.

Conclusion

Portfolio optimization is the "smart planning" part of investing. It is not just about picking the best-looking asset but about understanding how these assets behave together. Through diversification and correlation advantages, it is possible to achieve lower risk at the same return target or higher returns at the same risk. Even in a simple 3-asset example, we saw that equity-bond-gold combinations significantly reduce volatility.

In practice, setting clear goals, updating data regularly, accounting for transaction costs and taxes, and conducting disciplined rebalancing are the keys to success. In dynamic markets like Turkey, where currency, inflation, and interest rate regimes can shift rapidly, periodically re-"optimizing" your portfolio and staying flexible is particularly important.

• Modern Portfolio Theory: Fundamentals and the Efficient Frontier
• Risk Management in Investing: Volatility, VaR, and CVaR
• Fund Selection on TEFAS: Expense Ratio, Liquidity, and Tracking Error
• What Is an ETF? Advantages, Risks, and Selection Criteria
• How to Rebalance: A Practical Guide
• Tax Guide for Investors: Funds, Equities, Bonds, and Gold

Related articles: Portfolio Management Strategy Guide, Modern Portfolio Theory, What Is Diversification?, When to Rebalance Your Portfolio, Correlation in Portfolios.