Factor investing

Factor investing is an investment approach that involves selecting and weighting securities according to specific characteristics, known as factors, that are associated with differences in expected returns, risk, or other investment outcomes.

Factor investing can be implemented across asset classes and is commonly used in portfolio management, quantitative investing, and asset allocation.

Concept

A factor is a measurable characteristic of an asset or group of assets that can be used to explain differences in investment returns or risk.

A simplified factor-investing framework is:

$R_i={\alpha }_i+\sum _{k=1}^K{\beta }_{ik}{F}_k+{ϵ}_i$

where:

• $R_i$ = return of asset $i$;
• ${\alpha }_i$ = asset-specific component;
• ${\beta }_{ik}$ = sensitivity of asset $i$ to factor $k$;
• $F_k$ = return associated with factor $k$;
• ${ϵ}_i$ = residual component;
• $K$ = number of factors.

Factor investing seeks to obtain systematic exposure to one or more such factors through portfolio construction.

Major factors

Common equity factors include:

• value;
• momentum;
• quality;
• size;
• low volatility.

The exact definition of each factor varies among researchers, index providers, and investment managers.

Value factor

The value factor refers to the tendency of relatively inexpensive securities, measured using valuation metrics, to exhibit different long-term return characteristics from relatively expensive securities.

Common valuation measures include:

• price-to-earnings ratio;
• price-to-book ratio;
• enterprise-value-to-EBITDA ratio;
• dividend yield;
• earnings yield.

A simplified value characteristic can be represented as:

$\text{Earnings Yield}=\frac{\text{Earnings}}{\text{Market Price}}$

A higher earnings yield generally corresponds to a lower valuation relative to earnings.

Momentum factor

The momentum factor is associated with the tendency of securities that have recently performed relatively well or poorly to exhibit persistence in their relative performance over certain measurement periods.

A simplified momentum measure is:

$\text{Momentum Return}=R_{t-k,t}$

where $R_{t-k,t}$ represents the return over a specified historical period.

Momentum strategies typically rank securities according to their past returns and construct portfolios based on those rankings.

Quality factor

The quality factor refers to characteristics associated with the financial strength, profitability, stability, or efficiency of companies.

Potential quality measures include:

• return on equity;
• return on assets;
• profitability;
• earnings stability;
• leverage;
• cash-flow quality.

For example:

Failed to parse (syntax error): {\displaystyle \text{Return on Equity} = \frac{\text{Net Income}} {\text{Average Shareholders' Equity}} \] The specific measures used to define quality vary among factor methodologies. ==Size factor== The '''size factor''' is associated with differences between smaller and larger companies. A simplified measure of company size is market capitalization: <math> \text{Market Capitalization} = \text{Share Price} \times \text{Shares Outstanding} \] The size factor is commonly associated with the historical observation that smaller companies have, under some methodologies and periods, exhibited different risk and return characteristics from larger companies. ==Low-volatility factor== The '''low-volatility factor''' involves selecting securities with relatively low historical or expected volatility. A simplified volatility measure is the standard deviation of returns: <math> \sigma(R_i) = \sqrt{ \frac{1}{n-1} \sum_{t=1}^{n} (R_{i,t}-\bar{R}_i)^2 } \] A low-volatility strategy generally ranks or selects securities according to their volatility and constructs a portfolio with relatively lower volatility characteristics. ==Multifactor investing== Investors can combine multiple factors in a single portfolio. A simplified multifactor score can be represented as: <math> S_i = \sum_{k=1}^{K} w_kF_{ik} }

where:

• $S_i$ = composite factor score for asset $i$;
• $F_{ik}$ = exposure of asset $i$ to factor $k$;
• $w_k$ = weight assigned to factor $k$;
• $K$ = number of factors.

Portfolio weights can then be determined according to the resulting factor scores and the constraints of the investment strategy.

Factor models

Factor investing is closely related to factor models, which attempt to explain asset returns through systematic exposures.

A general factor model is:

$R_i={\alpha }_i+\sum _{k=1}^K{\beta }_{ik}{F}_k+{ϵ}_i$

Factor models can be used to estimate the sensitivity of securities to different systematic factors.

Arbitrage pricing theory

Arbitrage pricing theory (APT) provides a theoretical framework in which expected returns can depend on exposure to multiple systematic risk factors.

A simplified APT equation is:

$E(R_i)=R_f+\sum _{k=1}^K{\beta }_{ik}{\lambda }_k$

where:

• $E(R_i)$ = expected return of asset $i$;
• $R_f$ = risk-free rate;
• ${\beta }_{ik}$ = sensitivity to factor $k$;
• ${\lambda }_k$ = risk premium associated with factor $k$.

Factor investing can use similar concepts of systematic exposure, although practical factor-investing strategies do not necessarily represent implementations of APT.

Factor investing and CAPM

The Capital asset pricing model (CAPM) uses market beta as its principal systematic risk factor:

$E(R_i)=R_f+{\beta }_i[E(R_m)-R_f]$

Factor investing generally uses a broader set of characteristics or systematic exposures.

A simplified distinction is:

$\text{CAPM}\to \text{Market Factor}$

$\text{Multifactor Investing}\to \text{Multiple Factors}$

Factor exposure

Factor exposure represents the degree to which a security or portfolio is sensitive to a particular factor.

For a portfolio:

${\beta }_{pk}=\sum _{i=1}^n{w}_i{\beta }_{ik}$

where:

• ${\beta }_{pk}$ = portfolio exposure to factor $k$;
• $w_i$ = portfolio weight of asset $i$;
• ${\beta }_{ik}$ = asset $i$'s exposure to factor $k$.

This allows investors to measure the factor characteristics of an overall portfolio rather than only individual securities.

Portfolio construction

Factor-investing strategies can construct portfolios using methods such as:

• ranking securities by factor characteristics;
• selecting securities above or below a factor threshold;
• weighting securities according to factor scores;
• optimizing factor exposures;
• combining multiple factor portfolios.

Portfolio weights may be subject to constraints involving:

• diversification;
• sector exposure;
• turnover;
• liquidity;
• maximum security weights;
• transaction costs.

Factor diversification

Combining several factors can diversify exposure to individual factor cycles.

A simplified multifactor portfolio return can be represented as:

$R_p={\alpha }_p+\sum _{k=1}^K{\beta }_{pk}{F}_k+{ϵ}_p$

The interaction between factors can affect the overall risk and return characteristics of the portfolio.

Factor investing and passive management

Factor investing is often implemented through exchange-traded funds (ETFs) and index funds that follow rules-based factor indexes.

These strategies are sometimes described as smart beta because they combine elements of passive index construction with systematic selection or weighting rules.

Factor investing and active management

Factor investing can also be used as part of active management.

An active manager may deliberately increase or reduce portfolio exposure to selected factors based on expectations about their future performance.

The resulting portfolio can have active return relative to a benchmark:

$\text{Active Return}=R_p-R_b$

and the variability of those active returns can be measured by tracking error.

Factor performance

Factor performance can be evaluated by comparing the return of a factor portfolio with an appropriate benchmark.

A simplified factor return is:

$R_F=R_{\text{Factor Portfolio}}-R_{\text{Reference Portfolio}}$

The precise construction of factor returns depends on the factor definition and methodology.

Risk considerations

Factor strategies can experience periods of underperformance.

Potential risks include:

• factor cyclicality;
• concentration;
• unintended sector exposure;
• turnover;
• transaction costs;
• liquidity constraints;
• model risk;
• changes in factor relationships;
• crowded positioning.

Historical factor premiums do not guarantee future factor performance.

Importance

Factor investing provides a systematic approach to portfolio construction by targeting measurable characteristics associated with investment returns or risk. It provides a framework for implementing exposures to value, momentum, quality, size, low volatility, and other factors through rules-based portfolios and multifactor strategies.

See also

• Asset pricing
• Arbitrage pricing theory
• Capital asset pricing model
• Factor model
• Value investing
• Momentum investing
• Quality investing
• Size effect
• Low-volatility investing
• Smart beta
• Portfolio management
• Active management
• Passive management
• Risk-adjusted return
• Active return
• Tracking error
• Risk premium
• Beta (finance)
• Systematic risk
• Exchange-traded fund
• Index fund