How Psychology Shapes Quantitative Finance Models

zsinkala , , , ,

Integrating Investor Psychology with Mathematics in Quantitative Behavioral Finance

Quantitative Behavioral Finance (QBF) combines psychological insights with mathematical models to simulate real-world behavior, particularly focusing on biases such as overconfidence, loss aversion, and anchoring. By quantifying these biases, QBF can predict market movements influenced by psychological tendencies. Here’s a step-by-step breakdown of how each of these biases can be mathematically modeled.

1. Overconfidence Bias

Overconfidence leads investors to overestimate their knowledge or ability to predict market outcomes. This often results in excessive trading or underestimation of risks. Here’s how it can be mathematically represented:

Step 1: Define the Expected Return Model with Confidence Levels

Let R represent the expected return of an asset and C represent the investor’s confidence level. Overconfident investors may assign higher probabilities to successful outcomes than actual market probabilities suggest.

Step 2: Adjust Expected Return Based on Overconfidence

Overconfident investors perceive a higher expected return:

R' = μ + α * σ

where μ is the market return, α represents overconfidence, and σ is the standard deviation of returns. This formula shows that overconfident investors expect higher returns than the market average.

Step 3: Simulate Trading Volume

Overconfidence often increases trading volume. If T is the baseline trading volume, then overconfident trading volume T’ can be modeled as:

T' = T * (1 + β)

where β represents overconfidence. Higher values of β indicate greater overconfidence and higher trading volume.

2. Loss Aversion Bias

Loss aversion suggests that investors feel the pain of losses more intensely than the pleasure from equivalent gains. This can be represented using a modified utility function.

Step 1: Define the Utility Function with Loss Aversion

A loss-averse utility function can be defined as:

U(x) = x^γ, if x >= 0; -λ * (-x)^γ, if x < 0

where γ represents risk aversion, and λ (typically >1) represents the degree of loss aversion.

Step 2: Calculate Risk Aversion Adjusted for Loss Aversion

For example, if λ = 2, losses are felt twice as strongly as gains. This affects decision-making, causing investors to avoid risks even when expected returns are positive.

Step 3: Apply to Investment Decisions

For an investment with potential gain G and potential loss L, the net utility is calculated as:

Net Utility = U(G) + U(-L)

If the net utility is positive, the investor may consider the investment. However, loss aversion can lead to lower net utility and risk aversion, even when potential gains are substantial.

3. Anchoring Bias

Anchoring occurs when investors overly rely on a reference point, like a recent high price, when making decisions. Here’s how to model it:

Step 1: Define an Anchoring Reference Point

Let P₀ be an anchor point, such as a past high price. Investors’ perceived value of the asset may be influenced by P₀, even if the current market price P differs.

Step 2: Calculate Perceived Value with Anchoring

The perceived value P’ can be influenced by the anchor:

P' = θ * P₀ + (1 - θ) * P

where θ (0 < θ < 1) represents sensitivity to the anchor. A higher θ means more reliance on the anchor.

Step 3: Determine Investment Behavior Based on Anchored Perception

If P’ (the perceived value) is above the current price P, the investor may see the asset as undervalued. Conversely, if P’ is below P, they might view it as overvalued, even if fundamentals suggest otherwise.

4. Integrating These Biases into QBF Models

Each of these biases—overconfidence, loss aversion, and anchoring—can be integrated into predictive models and simulations:

  • Overconfidence: Overconfidence-adjusted returns lead to more aggressive portfolios.
  • Loss Aversion: Loss aversion increases the weight of low-risk assets in optimization models.
  • Anchoring: Anchoring affects price expectations, adjusting them gradually toward market reality.

By incorporating these psychological biases into mathematical models, QBF provides more realistic predictions of market behavior and investor decisions. The integration of investor psychology with mathematics offers a comprehensive approach to understanding market dynamics, ultimately leading to better-informed investment strategies and improved risk management.