Portfolio Construction Glossary
Plain-English definitions of key terms used in portfolio optimization, retirement planning, and asset allocation. Each term links to the relevant Portfolio Lab tool or research.
Arithmetic vs Geometric Return
Arithmetic return is the simple average of periodic returns. Geometric return (CAGR) accounts for compounding and is always lower. For portfolio planning, geometric return is what you actually earn over time. The difference between the two grows with volatility, which is why high-volatility assets like Bitcoin have a larger gap between their arithmetic and geometric returns.
Backtesting
Backtesting applies a portfolio allocation strategy to historical market data to see how it would have performed. It reveals how a portfolio would have behaved through real crashes, recoveries, and regime changes. Backtesting is useful for stress-testing but has limitations: past market conditions may not repeat.
Black-Litterman Model
A portfolio optimization method that starts from market equilibrium returns (what the market implies is optimal) and blends in your personal views with a specified confidence level. Developed by Fischer Black and Robert Litterman at Goldman Sachs in 1992, it solves the biggest practical problem with mean-variance optimization: extreme sensitivity to return estimates.
CAPE Ratio (Cyclically Adjusted P/E)
The CAPE ratio divides a stock market's current price by its average inflation-adjusted earnings over the past 10 years. Developed by Robert Shiller, it smooths out business cycle effects to give a longer-term view of whether a market is cheap or expensive relative to its own history. Higher CAPE ratios are associated with lower future returns.
Capital Market Assumptions (CMAs)
Forward-looking estimates of expected return, volatility, and correlation for each asset class over a long-term horizon (typically 10-15 years). Major research firms like J.P. Morgan, BlackRock, and Research Affiliates publish annual CMAs. Institutional investors use them as inputs for strategic asset allocation instead of relying on historical returns.
Conditional Value at Risk (CVaR / Expected Shortfall)
CVaR measures the average loss in the worst-case scenarios beyond the VaR threshold. While VaR tells you 'losses won't exceed X with 95% confidence,' CVaR tells you 'when losses do exceed X, the average loss is Y.' CVaR is considered a more complete measure of tail risk than VaR alone.
Correlation
A statistical measure ranging from -1 to +1 that describes how two assets move relative to each other. A correlation of +1 means they move in lockstep, -1 means they move in opposite directions, and 0 means no relationship. Low or negative correlations between assets are the foundation of portfolio diversification.
Dollar-Cost Averaging (DCA)
An investment strategy where you invest a fixed amount at regular intervals regardless of price. DCA reduces the impact of volatility by automatically buying more shares when prices are low and fewer when prices are high. It removes the need to time the market and is psychologically easier for volatile assets like Bitcoin.
Drawdown
The peak-to-trough decline in a portfolio's value before it recovers to a new high. Maximum drawdown measures the largest such decline over a given period. Drawdowns are a critical risk measure because they represent the actual pain investors experience, and recovery time can be substantial even when average returns are positive.
Efficient Frontier
The set of portfolios that offer the highest expected return for each level of risk, or equivalently, the lowest risk for each level of return. Any portfolio below the frontier is suboptimal because you could achieve better returns at the same risk. The efficient frontier is the core output of mean-variance optimization.
Factor Exposure
The degree to which a portfolio's returns are driven by systematic risk factors such as value, momentum, quality, size, and low volatility. Understanding factor exposures reveals whether your portfolio's risk is truly diversified or concentrated in one or two factors, even if you hold many different assets.
Hierarchical Risk Parity (HRP)
A machine learning-based portfolio optimization method developed by Marcos Lopez de Prado in 2016. HRP clusters assets by how similarly they behave, then allocates within and across clusters using inverse-variance weighting. Unlike traditional optimization, HRP does not require expected return estimates and avoids inverting the covariance matrix, making it more robust to estimation error.
Mean-Variance Optimization (MVO)
The foundational portfolio construction technique developed by Harry Markowitz in 1952. MVO finds the combination of asset weights that maximizes expected return for a given level of risk (standard deviation) or minimizes risk for a given return target. It requires estimates of expected returns, volatilities, and correlations for all assets.
Monte Carlo Simulation
A technique that generates thousands of random future scenarios for a portfolio by sampling from a statistical model of asset returns. Instead of projecting a single average outcome, it produces a distribution of possible outcomes with probabilities. Monte Carlo simulation is the standard method for retirement planning because it captures sequence risk and tail events.
Rebalancing
The process of restoring a portfolio to its target asset allocation by selling assets that have grown above their target weight and buying those that have fallen below it. Rebalancing maintains the intended risk profile and can improve risk-adjusted returns through a disciplined sell-high, buy-low mechanism.
Risk Parity
A portfolio construction method that allocates assets so that each contributes equally to total portfolio risk, rather than equalizing dollar amounts. Risk parity typically results in higher bond allocations than traditional portfolios because bonds are less volatile. It uses only volatilities and correlations as inputs, requiring no expected return estimates.
Sequence Risk (Sequence of Returns Risk)
The risk that poor market returns occur early in retirement when the portfolio is at its largest and most vulnerable to withdrawals. Two retirees with identical average returns can have vastly different outcomes depending on the order of those returns. Sequence risk is the primary reason why safe withdrawal rates are lower than average expected returns.
Safe Withdrawal Rate (SWR)
The highest percentage of a retirement portfolio that can be withdrawn annually (adjusted for inflation) without running out of money over a specified time horizon. The original Trinity Study (1998) found 4% to be safe over 30 years using historical US data. Forward-looking analysis using current assumptions suggests the safe rate may be lower.
Value at Risk (VaR)
A risk metric that estimates the maximum loss a portfolio could experience over a given period at a specified confidence level. For example, a 95% VaR of 12% means there is a 5% chance the portfolio could lose more than 12% in that period. VaR is widely used by institutional investors and regulators to quantify downside risk.
Volatility (Standard Deviation)
A statistical measure of how much an asset's returns vary from their average. Higher volatility means larger price swings in both directions. In portfolio construction, volatility is the standard proxy for risk. Annualized volatility is calculated as the standard deviation of periodic returns multiplied by the square root of the number of periods per year.
See these concepts in action
Every term in this glossary is implemented in a free Portfolio Lab tool.