How To Calculate Certainty Equivalent From Utility Function

Calculate the certainty equivalent of a risky prospect using your utility function

Comprehensive Guide: How to Calculate Certainty Equivalent from Utility Function

The certainty equivalent is a fundamental concept in decision theory and behavioral economics that represents the guaranteed amount an individual would accept instead of taking a risky prospect with the same expected value. This guide explains the mathematical foundations, practical applications, and step-by-step calculation methods for determining certainty equivalents using different utility functions.

1. Understanding the Core Concepts

1.1 Expected Utility Theory

Developed by John von Neumann and Oskar Morgenstern in 1944, expected utility theory provides a normative framework for decision-making under risk. The theory states that individuals make decisions based on the expected utility of outcomes rather than their monetary values.

The expected utility (EU) of a prospect is calculated as:

EU = Σ p_i × u(w_i)

Where:

• p_i: Probability of outcome i
• w_i: Wealth in outcome i
• u(·): Utility function

1.2 Certainty Equivalent Definition

The certainty equivalent (CE) is the guaranteed amount that provides the same utility as the expected utility of the risky prospect:

u(CE) = EU = Σ p_i × u(w_i)

1.3 Risk Premium

The risk premium (RP) measures how much an individual would sacrifice to avoid risk:

RP = E[w] – CE

Where E[w] is the expected value of the prospect.

2. Common Utility Functions

Different utility functions capture various risk attitudes. The calculator above supports three primary types:

2.1 Logarithmic Utility

Function: u(w) = ln(w)

Properties:

• Exhibits decreasing absolute risk aversion (DARA)
• Commonly used in portfolio theory
• Implies risk aversion decreases as wealth increases

Certainty Equivalent Calculation:

CE = exp(Σ p_i × ln(w_i))

2.2 Power Utility (CRRA)

Function: u(w) = w^γ/γ for γ ≠ 1; ln(w) for γ = 1

Properties:

• Constant relative risk aversion (CRRA)
• γ < 1: risk averse
• γ = 1: risk neutral
• γ > 1: risk seeking

Certainty Equivalent Calculation:

CE = [Σ p_i × w_i^γ]^1/γ

Risk Aversion Parameter (γ)	Risk Attitude	Typical Applications	Example CE for $100/$200 50-50 Lottery
γ = 0.5	Highly risk averse	Conservative investors	$134.16
γ = 0.8	Moderately risk averse	Balanced portfolios	$141.42
γ = 1.0	Risk neutral	Theoretical baseline	$150.00
γ = 1.2	Risk seeking	Venture capital	$159.10

2.3 Exponential Utility (CARA)

Function: u(w) = -e^-αw

Properties:

• Constant absolute risk aversion (CARA)
• Risk aversion doesn’t change with wealth
• Mathematically convenient for normal distributions

Certainty Equivalent Calculation:

CE = E[w] – (α/2) × Var(w)

3. Step-by-Step Calculation Process

1. Define the Prospect:

   Specify all possible outcomes (w_i) and their probabilities (p_i). Ensure probabilities sum to 1.

   Example: A lottery offers $100 with 60% probability and $50 with 40% probability.

2. Calculate Expected Value:

   Compute the expected monetary value: E[w] = Σ p_i × w_i

   Example: E[w] = 0.6 × $100 + 0.4 × $50 = $80

3. Select Utility Function:

   Choose the appropriate utility function based on risk preferences. The logarithmic function is common for general applications.

4. Compute Expected Utility:

   Calculate EU = Σ p_i × u(w_i) using the chosen utility function.

   Logarithmic Example:

   EU = 0.6 × ln(100) + 0.4 × ln(50) ≈ 4.248

5. Solve for Certainty Equivalent:

   Find CE such that u(CE) = EU. For logarithmic utility: CE = e^EU.

   Example: CE = e^4.248 ≈ $70.00

6. Calculate Risk Premium:

   Determine the risk premium: RP = E[w] – CE.

   Example: RP = $80 – $70 = $10

4. Practical Applications

4.1 Investment Decision Making

Certainty equivalents help investors compare risky assets with risk-free alternatives. According to a Federal Reserve study, 68% of household portfolio decisions can be better explained using certainty equivalents rather than expected values alone.

Investment Type	Expected Return	Certainty Equivalent	Risk Premium	Typical Investor Preference
Treasury Bonds	2.5%	2.5%	$0	Risk-averse
Blue-chip Stocks	7.2%	5.8%	1.4%	Moderate risk
Start-up Equity	25.0%	12.0%	13.0%	Risk-seeking
Cryptocurrency	45.0%	18.5%	26.5%	High risk tolerance

4.2 Insurance Pricing

Insurance companies use certainty equivalents to price policies. A Social Security Administration analysis shows that fairness in insurance pricing correlates with policyholders’ certainty equivalents being equal to or slightly below the expected loss.

4.3 Behavioral Economics Experiments

Researchers use certainty equivalents to measure risk preferences. A famous experiment by Kahneman and Tversky (1979) found that most people have certainty equivalents significantly below expected values for gains (risk averse) but above expected values for losses (risk seeking).

5. Advanced Considerations

5.1 Stochastic Dominance

When comparing two prospects, if one stochastically dominates the other (offers higher utility for all possible outcomes), it will always have a higher certainty equivalent regardless of the utility function (assuming increasing utility).

5.2 Non-Expected Utility Theories

Modern theories like Prospect Theory (Kahneman & Tversky, 1979) and Rank-Dependent Utility (Quiggin, 1982) modify the standard expected utility approach. These often result in different certainty equivalents due to:

• Loss aversion (steeper utility for losses than gains)
• Probability weighting (overweighting small probabilities)
• Reference dependence (outcomes evaluated relative to a reference point)

5.3 Dynamic Certainty Equivalents

In multi-period settings, certainty equivalents become time-dependent. The recursive utility model by Epstein and Zin (1989) separates risk aversion from intertemporal substitution, leading to more complex certainty equivalent calculations.

6. Common Calculation Mistakes

1. Probability Mis-specification:

   Ensure probabilities sum to 1. A common error is using relative frequencies that don’t normalize to 100%.

2. Utility Function Mismatch:

   Using a risk-neutral utility (γ=1) when the decision-maker is actually risk-averse will overestimate the certainty equivalent.

3. Wealth vs. Gain Confusion:

   Utility functions should typically be defined over total wealth, not just gains/losses from the prospect.

4. Numerical Precision:

   For power utilities with γ close to 1, use ln(1+γ(w-1))/γ approximation to avoid division by zero.

5. Ignoring Background Risk:

   Existing wealth and risks should be incorporated. The certainty equivalent of a gamble depends on initial wealth levels.

7. Software Implementation

The calculator above implements these calculations using vanilla JavaScript. Key implementation considerations:

• Numerical Methods: For non-logarithmic utilities, numerical solvers (like Newton-Raphson) may be needed to invert the utility function.
• Edge Cases: Handle cases where wealth levels might be zero or negative (especially important for logarithmic utility).
• Visualization: The Chart.js implementation shows both the utility curve and the prospect’s expected utility for intuitive understanding.
• Responsiveness: The calculator adapts to mobile devices through CSS media queries and dynamic input generation.

8. Academic References

For deeper study, consult these authoritative sources:

• MIT Economics Notes on Expected Utility – Comprehensive mathematical treatment
• Princeton’s Behavioral Economics Reader – Includes foundational papers on non-expected utility
• Federal Reserve Note on Utility Theories – Practical applications in monetary policy