Home

Topic 3: Statistical Measures of Asset Returns

16 min read

Statistical Measures of Asset Returns

Learning Objectives

By the end of this topic, candidates should be able to:

  1. Calculate, interpret, and evaluate measures of central tendency and location to address an investment problem
  2. Calculate, interpret, and evaluate measures of dispersion to address an investment problem
  3. Interpret and evaluate measures of skewness and kurtosis to address an investment problem
  4. Interpret correlation between two variables to address an investment problem

Learning Objective 1: Measures of Central Tendency and Location

Core Concept

Central tendency measures provide a single representative value that describes the “center” or typical value of a dataset. In finance, these measures help investors understand the expected or typical return of an investment, but each has different strengths and applications. The arithmetic mean serves as the default for estimating forward-looking expected returns (as used in portfolio optimization), while the geometric mean is more appropriate for evaluating historical compound growth — a distinction that builds directly on the return measurement concepts from Topic 1.

  • Arithmetic Mean: The simple average, most commonly used for expected returns
  • Median: The middle value, less sensitive to extreme observations
  • Mode: The most frequently occurring value
  • Geometric Mean: More appropriate for multi-period returns and growth rates
  • Weighted Mean: Used when observations have different importance
  • Percentiles and Quartiles: Divide the distribution into equal parts

Financial Interpretation: Central tendency measures help investors set return expectations, compare investments, and understand the typical performance of assets. The choice of measure depends on the distribution characteristics and the specific investment question.

Formulas & Calculations

Arithmetic Mean

Formula: X̄ = (ΣXᵢ) / n formula

HP 12C Calculation: hp12c

  1. Clear statistical registers: [f] [CLx]
  2. Enter each return: value [Σ+]
  3. Calculate mean: [g] [x̄]

Geometric Mean

Formula: G = [(1 + R₁)(1 + R₂)…(1 + Rₙ)]^(1/n) - 1 formula

HP 12C Calculation:

  1. Convert returns to growth factors: (1 + return)
  2. Use [yx] function for compound calculations
  3. Take nth root: n [1/x] [yx]

Weighted Mean

Formula: X̄w = Σ(wᵢ × Xᵢ) / Σwᵢ

HP 12C Calculation:

  1. Multiply each value by its weight
  2. Sum weighted values and weights separately
  3. Divide total weighted values by total weights

Percentiles and Quartiles

  • First Quartile (Q1): 25th percentile
  • Second Quartile (Q2): 50th percentile (median)
  • Third Quartile (Q3): 75th percentile

Practical Examples

Example 1: Portfolio Return Analysis

A portfolio manager tracks monthly returns for a balanced fund:

  • Returns: -2%, 3%, 1%, 4%, -1%, 2%, 5%, 0%, 3%, 1%

HP 12C Calculation:

[f] [CLx]          // Clear registers
-2 [Σ+]           // Enter -2%
3 [Σ+]            // Enter 3%
1 [Σ+]            // Enter 1%
4 [Σ+]            // Enter 4%
-1 [Σ+]           // Enter -1%
2 [Σ+]            // Enter 2%
5 [Σ+]            // Enter 5%
0 [Σ+]            // Enter 0%
3 [Σ+]            // Enter 3%
1 [Σ+]            // Enter 1%
[g] [x̄]           // Calculate mean = 1.6%

Financial Interpretation: The average monthly return is 1.6%, suggesting the fund generates positive returns on average, suitable for moderate-risk investors seeking steady growth.

Example 2: Comparing Growth Rates

Two tech stocks over 3 years:

  • Stock A: 15%, -5%, 25%
  • Stock B: 10%, 12%, 8%

Arithmetic vs Geometric Mean:

  • Stock A Arithmetic: (15% - 5% + 25%) / 3 = 11.67%
  • Stock A Geometric: [(1.15)(0.95)(1.25)]^(1/3) - 1 = 10.76%
  • Stock B Arithmetic: (10% + 12% + 8%) / 3 = 10%
  • Stock B Geometric: [(1.10)(1.12)(1.08)]^(1/3) - 1 = 9.99%

Financial Interpretation: Stock A has higher average returns but more volatility. The geometric mean better represents actual compound growth, showing Stock A’s volatility reduces its compound return advantage.

DeFi Application

Token Return Distribution Analysis

Analyzing daily returns for a governance token over 30 days: defi-application

Data Collection: Track daily price changes for popular DeFi tokens (e.g., UNI, AAVE, COMP)

Central Tendency Analysis:

  • Mean: Expected daily return for risk assessment
  • Median: More robust measure for heavily skewed crypto returns
  • Mode: Identify if certain return levels are common (often near zero)

Practical Application:

# Example DeFi token analysis
daily_returns = [-0.05, 0.12, -0.08, 0.15, -0.03, 0.07, -0.12, 0.09]
 
# Risk assessment using different measures
mean_return = np.mean(daily_returns)      # 0.01875 (1.875%)
median_return = np.median(daily_returns)  # 0.01 (1%)
geometric_mean = ((1 + pd.Series(daily_returns)).prod()) ** (1/len(daily_returns)) - 1

Financial Interpretation: DeFi tokens often show high volatility with positive skew. The median may be more representative than the mean for setting realistic expectations, while geometric mean better reflects actual portfolio growth in compound scenarios.

Learning Objective 2: Measures of Dispersion

Core Concept

Dispersion measures quantify the spread or variability of returns around the central tendency. In finance, dispersion is synonymous with risk — higher dispersion indicates higher uncertainty and potential for both gains and losses. Understanding dispersion is critical because it feeds directly into portfolio variance calculations, simulation models, and the risk-adjusted return metrics that drive investment decisions. exam-focus

Key measures include:

  • Range: Difference between maximum and minimum values
  • Mean Absolute Deviation (MAD): Average absolute difference from the mean
  • Variance: Average squared deviation from the mean
  • Standard Deviation: Square root of variance, expressed in original units
  • Coefficient of Variation: Standardized measure of relative risk

Financial Interpretation: These measures help investors quantify risk, compare the risk-return profiles of different investments, and make informed portfolio allocation decisions.

Formulas & Calculations

Population vs Sample Formulas

Population Variance: σ² = Σ(Xᵢ - μ)² / N formula Sample Variance: s² = Σ(Xᵢ - X̄)² / (n-1) formula

Population Standard Deviation: σ = √[Σ(Xᵢ - μ)² / N] formula Sample Standard Deviation: s = √[Σ(Xᵢ - X̄)² / (n-1)] formula

HP 12C Statistical Functions

Standard Deviation Calculation:

  1. Clear registers: [f] [CLx]
  2. Enter data points: value [Σ+] (repeat for each)
  3. Sample standard deviation: [g] [s]
  4. Population standard deviation: [g] [σ]

Mean Absolute Deviation

Formula: MAD = Σ|Xᵢ - X̄| / n

Coefficient of Variation

Formula: CV = s / X̄ (expressed as percentage) formula exam-focus

HP 12C Calculation:

  1. Calculate standard deviation: [g] [s]
  2. Calculate mean: [g] [x̄]
  3. Divide: standard deviation ÷ mean

Practical Examples

Example 1: Risk Comparison of Two Mutual Funds

Fund A monthly returns: 2%, 4%, 1%, 6%, 3% Fund B monthly returns: 1%, 8%, -2%, 7%, 1%

HP 12C Calculation for Fund A:

[f] [CLx]          // Clear registers
2 [Σ+]            // Enter 2%
4 [Σ+]            // Enter 4%
1 [Σ+]            // Enter 1%
6 [Σ+]            // Enter 6%
3 [Σ+]            // Enter 3%
[g] [x̄]           // Mean = 3.2%
[g] [s]           // Sample std dev = 1.92%

Fund A: Mean = 3.2%, Standard Deviation = 1.92% Fund B: Mean = 3.0%, Standard Deviation = 4.24%

Coefficient of Variation:

  • Fund A: CV = 1.92% / 3.2% = 0.60 (60%)
  • Fund B: CV = 4.24% / 3.0% = 1.41 (141%)

Financial Interpretation: Fund A offers better risk-adjusted returns (lower CV), making it more attractive for risk-averse investors despite similar average returns.

Example 2: Sharpe Ratio Application

Portfolio with 12% annual return, 8% risk-free rate, 15% standard deviation: practice-problem

Sharpe Ratio = (Rp - Rf) / σp = (12% - 8%) / 15% = 0.267 formula exam-focus

HP 12C Calculation:

12 [ENTER] 8 [-]   // Excess return = 4%
15 [÷]             // Divide by std dev = 0.267

Financial Interpretation: A Sharpe ratio of 0.267 indicates the portfolio generates 0.267 units of excess return per unit of risk. Higher Sharpe ratios indicate better risk-adjusted performance.

DeFi Application

Crypto Portfolio Risk Analysis

DeFi tokens often exhibit high volatility with periods of extreme price movements followed by calmer periods — a phenomenon known as “volatility clustering” that challenges the constant-variance assumption of many traditional models. defi-application

Example Analysis:

# DeFi token risk metrics
import pandas as pd
import numpy as np
 
# Sample daily returns for popular DeFi tokens
uni_returns = [0.05, -0.12, 0.08, -0.15, 0.22, -0.08, 0.03]
aave_returns = [0.03, -0.07, 0.12, -0.09, 0.18, -0.05, 0.01]
 
# Risk calculations
uni_std = np.std(uni_returns, ddof=1)  # Sample standard deviation
aave_std = np.std(aave_returns, ddof=1)
 
uni_cv = uni_std / np.mean(uni_returns)
aave_cv = aave_std / np.mean(aave_returns)

Practical Applications:

  1. Position Sizing: Use standard deviation to determine appropriate position sizes
  2. Risk Parity: Allocate capital based on inverse volatility weighting
  3. Stress Testing: Use historical volatility to model potential losses

Financial Interpretation: DeFi protocols often show much higher volatility than traditional assets (20-100% annualized vs 10-20% for stocks). This requires more sophisticated risk management and smaller position sizes for equivalent risk exposure.

Learning Objective 3: Measures of Skewness and Kurtosis

Core Concept

Skewness and kurtosis describe the shape characteristics of return distributions beyond central tendency and dispersion. These measures matter because real-world returns deviate from the normal distribution assumed by many models, and ignoring these deviations can lead to serious underestimation of risk. Understanding skewness and kurtosis becomes especially important when we move into simulation methods, where the choice of distribution directly affects the realism of results. exam-focus

  1. Investment returns are rarely normally distributed
  2. Extreme events (fat tails) are more common than normal distribution predicts
  3. Asymmetric distributions affect risk assessment and portfolio decisions

Skewness

Measures the asymmetry of the distribution:

  • Positive Skew (Right-skewed): Long right tail, mean > median
  • Negative Skew (Left-skewed): Long left tail, mean < median
  • Zero Skew: Symmetric distribution, mean ≈ median

Kurtosis

Measures the “tailedness” or peakedness:

  • Leptokurtic: Kurtosis > 3, fatter tails than normal distribution
  • Mesokurtic: Kurtosis = 3, normal distribution
  • Platykurtic: Kurtosis < 3, thinner tails than normal distribution

Financial Interpretation:

  • Positive skew generally preferred (more upside potential)
  • High kurtosis indicates higher probability of extreme events
  • Most financial assets exhibit negative skew and positive excess kurtosis

Formulas & Calculations

Skewness Formula

Sample Skewness: Sk = [n/((n-1)(n-2))] × [Σ(Xᵢ - X̄)³/s³]

Kurtosis Formula

Sample Kurtosis: K = [n(n+1)/((n-1)(n-2)(n-3))] × [Σ(Xᵢ - X̄)⁴/s⁴] - [3(n-1)²/((n-2)(n-3))]

HP 12C Approach

While HP 12C doesn’t directly calculate skewness and kurtosis, you can:

  1. Calculate mean and standard deviation using statistical functions
  2. Use the formula approach with manual calculations for higher moments
  3. Focus on qualitative assessment by examining distribution shape

Interpretation Guidelines

Skewness:

  • Sk > 0.5: Highly positively skewed
  • 0.5 > Sk > -0.5: Approximately symmetric
  • Sk < -0.5: Highly negatively skewed

Kurtosis:

  • K > 1: Leptokurtic (fat tails)
  • -1 < K < 1: Approximately normal
  • K < -1: Platykurtic (thin tails)

Practical Examples

Example 1: Comparing Stock Return Distributions

Two stocks over 12 months:

Stock A Returns: 2%, 3%, 1%, 4%, 2%, 8%, 15%, 3%, 2%, 4%, 1%, 5% Stock B Returns: 5%, -12%, 8%, 3%, 4%, 2%, 6%, -8%, 12%, 4%, 3%, 7%

Analysis Using HP 12C:

// Stock A Analysis
[f] [CLx]
// Enter all returns using [Σ+]
[g] [x̄]    // Mean = 4.17%
[g] [s]    // Std Dev = 4.26%

// Stock B Analysis  
[f] [CLx]
// Enter all returns using [Σ+]
[g] [x̄]    // Mean = 2.83%
[g] [s]    // Std Dev = 7.18%

Qualitative Assessment:

  • Stock A: Large positive outlier (15%) suggests positive skew
  • Stock B: Both large positive and negative outliers suggest higher kurtosis

Financial Interpretation: Stock A shows positive skew (desirable for investors), while Stock B shows higher volatility with potential for extreme events in both directions.

Example 2: Market Crisis Analysis

During market stress, returns often exhibit:

  • Negative skew: More frequent large losses than gains
  • High kurtosis: Extreme events more common than normal distribution predicts

2008 Financial Crisis Example:

  • S&P 500 daily returns showed significant negative skew (-1.5)
  • Excess kurtosis of +5, indicating fat tails
  • This contradicted models assuming normal distributions

DeFi Application

Analyzing DeFi Token Return Distributions

Typical DeFi Token Characteristics:

  1. High positive skew: Potential for explosive gains during bull markets
  2. Very high kurtosis: Extreme price movements common
  3. Volatility clustering: Periods of high volatility followed by calm

Example: Governance Token Analysis

# Example DeFi token daily returns analysis
import scipy.stats as stats
 
# Sample governance token returns (%)
token_returns = [0.05, -0.15, 0.25, -0.08, 0.45, -0.12, 0.08, -0.20, 0.35, -0.10]
 
# Calculate shape measures
skewness = stats.skew(token_returns)      # Measure of asymmetry
kurtosis = stats.kurtosis(token_returns)  # Measure of tail risk
 
print(f"Skewness: {skewness:.3f}")
print(f"Kurtosis: {kurtosis:.3f}")

Practical Risk Management Applications

  1. Position Sizing: Adjust for negative skew and high kurtosis

    • Smaller positions in negatively skewed assets
    • Account for tail risk in position sizing models

  2. Options Strategies:

    • High kurtosis suggests options may be underpriced
    • Positive skew favors call options, negative skew favors puts

  3. Liquidity Provision:

    • High kurtosis increases impermanent loss risk
    • Adjust fee parameters based on expected distribution shape

Financial Interpretation: DeFi tokens often show extreme distribution characteristics (high skew and kurtosis) requiring sophisticated risk management. Traditional portfolio theory may underestimate risks, necessitating stress testing and scenario analysis.

Learning Objective 4: Correlation Between Two Variables

Core Concept

Correlation measures the linear relationship between two variables, ranging from -1 to +1. In finance, correlation is fundamental to portfolio construction and risk management. The diversification benefit of combining assets into a portfolio depends entirely on the correlation structure — this is one of the central insights of Modern Portfolio Theory.

  • Portfolio diversification: Lower correlations reduce portfolio risk
  • Risk management: Understanding how assets move together
  • Hedging strategies: Using negatively correlated assets for protection
  • Factor analysis: Identifying common risk factors

Key Concepts:

  • Correlation ≠ Causation: Statistical relationship doesn’t imply one causes the other
  • Linear relationship only: Correlation may miss non-linear relationships
  • Time-varying: Correlations change over time, especially during crises
  • Sample correlation: Estimate based on historical data

Financial Interpretation: Correlation drives diversification benefits. Perfect positive correlation (+1) provides no diversification, while negative correlation (-1) offers maximum diversification benefits.

Formulas & Calculations

Correlation Coefficient Formula

Population: ρ(X,Y) = Cov(X,Y) / (σX × σY) formula Sample: r = Σ[(Xᵢ - X̄)(Yᵢ - Ȳ)] / √[Σ(Xᵢ - X̄)² × Σ(Yᵢ - Ȳ)²] formula

Covariance Formula

Sample Covariance: Cov(X,Y) = Σ[(Xᵢ - X̄)(Yᵢ - Ȳ)] / (n-1) formula

HP 12C Calculation Method

While HP 12C doesn’t directly calculate correlation, you can use the regression functions:

  1. Enter paired data using [Σ+]
  2. The correlation coefficient is related to the regression statistics
  3. Use: r = √(coefficient of determination)

Alternative HP 12C Approach:

  1. Calculate means of both variables separately
  2. Calculate deviations and products manually
  3. Use statistical functions for sums and standard deviations

Interpretation Guidelines

  • r > 0.7: Strong positive correlation
  • 0.3 < r < 0.7: Moderate positive correlation
  • -0.3 < r < 0.3: Weak correlation
  • -0.7 < r < -0.3: Moderate negative correlation
  • r < -0.7: Strong negative correlation

Practical Examples

Example 1: Stock-Bond Correlation Analysis

Monthly returns for stock index and government bonds:

Stock Returns: 2%, -3%, 5%, 1%, -2%, 4%, 3%, -1%, 2%, 0% Bond Returns: -1%, 2%, -2%, 1%, 3%, -1%, -2%, 2%, 0%, 1%

HP 12C Calculation:

[f] [CLx]           // Clear statistical registers
2 [ENTER] -1 [Σ+]   // Enter first pair (Stock: 2%, Bond: -1%)
-3 [ENTER] 2 [Σ+]   // Enter second pair  
5 [ENTER] -2 [Σ+]   // Continue for all pairs...
// Continue entering all paired data
[g] [ŷ,r]          // This gives correlation coefficient

Expected Result: Negative correlation (~-0.4), showing diversification benefits.

Financial Interpretation: The negative correlation indicates that when stocks decline, bonds tend to rise, providing portfolio protection during market stress.

Example 2: Sector Correlation in Portfolio Construction

Technology vs. Utilities sector returns:

Analysis for Portfolio Allocation:

  • High positive correlation (>0.8): Limited diversification benefit
  • Low correlation (<0.3): Good diversification opportunity
  • Negative correlation: Excellent hedging potential

Example 3: Currency Hedging Decision

US investor holding European stocks needs to assess EUR/USD correlation with European equity returns:

  • Positive correlation: Currency and stocks move together, increasing risk
  • Negative correlation: Natural hedge, currency losses offset by stock gains
  • Zero correlation: Currency and stock risks independent

DeFi Application

Cross-Protocol Correlation Analysis

Understanding correlations across DeFi protocols is essential for constructing diversified yield farming strategies. During normal conditions, protocols may show moderate correlation, but during stress events (e.g., a stablecoin depeg or a major exploit), correlations tend to spike toward 1 — eliminating diversification benefits exactly when they are most needed. defi-application

DeFi Ecosystem Correlations:

  1. Governance tokens: Often highly correlated during market moves
  2. Stablecoin yields: May correlate with broader DeFi adoption
  3. Layer 1 vs DeFi tokens: Varying correlations based on market conditions

Example: Yield Farming Strategy

Analyzing correlation between different DeFi protocol tokens for optimal farm diversification:

# Example correlation analysis for DeFi portfolio
import numpy as np
import pandas as pd
 
# Daily returns for major DeFi protocols
uni_returns = [0.05, -0.12, 0.08, -0.03, 0.15]
aave_returns = [0.03, -0.08, 0.12, -0.01, 0.09]
comp_returns = [0.07, -0.15, 0.05, -0.02, 0.18]
 
# Create correlation matrix
returns_df = pd.DataFrame({
    'UNI': uni_returns,
    'AAVE': aave_returns, 
    'COMP': comp_returns
})
 
correlation_matrix = returns_df.corr()
print("DeFi Protocol Correlations:")
print(correlation_matrix)

Practical Applications in DeFi:

  1. Liquidity Mining: Diversify across protocols with low correlations
  2. Impermanent Loss Management: Consider correlation between paired tokens
  3. Cross-chain Arbitrage: Exploit correlation differences across chains
  4. Risk Management: Monitor correlation increases during market stress

Correlation Breakdown During Crises

Key Insight: Correlations tend to increase during market stress, reducing diversification benefits when most needed.

Example: During March 2020 crypto crash:

  • Most DeFi tokens showed correlations >0.9 with Bitcoin
  • Traditional diversification strategies failed
  • Only stablecoins and some derivatives maintained low correlations

Financial Interpretation: DeFi correlations are highly time-varying and tend toward 1 during market stress. This requires dynamic hedging strategies and stress testing of correlation assumptions in portfolio models.

Comprehensive Review and Applications

Risk-Adjusted Performance Metrics

Sharpe Ratio

Formula: Sharpe Ratio = (Rp - Rf) / σp

HP 12C Calculation:

Portfolio_Return [ENTER] Risk_Free_Rate [-]
Portfolio_StdDev [÷]

Sortino Ratio (Downside Deviation)

Focuses on downside risk only, more relevant for investors concerned about losses.

Information Ratio

Measures active return per unit of tracking error for actively managed portfolios.

Portfolio Applications

Modern Portfolio Theory Integration

  • Expected Return: Use arithmetic mean for single-period analysis
  • Risk: Use standard deviation or variance
  • Correlation: Essential for portfolio optimization
  • Skewness/Kurtosis: Adjust for non-normal distributions

Risk Budgeting

Allocate risk based on:

  1. Individual asset volatility
  2. Correlation structure
  3. Expected skewness and kurtosis
  4. Tail risk measures

DeFi-Specific Considerations

Unique Risk Characteristics

  1. Smart Contract Risk: Additional source of correlation during protocol failures
  2. Liquidity Risk: Correlations may spike during liquidity crises
  3. Governance Risk: Token holder decisions can create sudden correlation changes
  4. Regulatory Risk: Affects entire DeFi ecosystem simultaneously

Advanced Applications

  1. Dynamic Hedging: Adjust positions based on changing correlations
  2. Yield Optimization: Consider risk-adjusted yields using Sharpe ratios
  3. Impermanent Loss Modeling: Use correlation and volatility to estimate losses
  4. Cross-chain Analysis: Compare risk-return profiles across different blockchains

Key Takeaways for Finance Certification 1 exam-focus

Essential Concepts to Remember

  1. Central tendency measures each serve different purposes in financial analysis
  2. Dispersion measures quantify risk and enable risk-adjusted comparisons
  3. Skewness and kurtosis reveal distribution characteristics not captured by mean and standard deviation
  4. Correlation drives diversification benefits but may fail during crises
  5. HP 12C statistical functions provide efficient calculation methods for all basic measures

Common Exam Applications

  • Calculate and interpret portfolio risk and return metrics
  • Compare investment alternatives using risk-adjusted measures
  • Assess diversification benefits using correlation analysis
  • Evaluate distribution characteristics for risk assessment
  • Apply statistical measures to investment decision-making scenarios

Financial Interpretation Focus

Always connect statistical calculations to financial decision-making:

  • How do these measures inform investment choices?
  • What are the implications for portfolio construction?
  • How do distribution characteristics affect risk assessment?
  • When might traditional measures be misleading?

The key to success in Finance Certification 1 is not just memorizing formulas, but understanding how statistical measures inform real-world investment decisions and risk management strategies.

Graph View

Backlinks