Understanding Correlation Across Asset Classes

This sub‑topic explains the concept of correlation across asset classes, how it is measured, and why it matters for portfolio risk management. Understanding correlation helps a PMS distributor recommend diversified portfolios that meet SEBI risk‑management expectations. The content links directly to the Portfolio Management Process chapter and is frequently tested in the NISM Series XXI‑A exam.

Learning Objectives

1Define correlation and differentiate it from causation.
2Calculate the Pearson correlation coefficient using the official formula.
3Interpret typical correlation levels among Indian asset classes.
4Apply correlation to assess portfolio diversification and variance.

What is Correlation?

Correlation measures the statistical relationship between the returns of two assets. It is expressed as a number between –1 and +1, where +1 indicates perfect positive movement, –1 indicates perfect inverse movement, and 0 indicates no linear relationship.

In the context of portfolio management, correlation tells a distributor whether combining two asset classes will reduce overall portfolio risk. A low or negative correlation between assets means that when one asset falls, the other may rise or stay stable, cushioning the portfolio.

Exam candidates often see correlation questions framed as: “If equities and debt have a correlation of 0.25, what is the impact on portfolio variance?” Knowing the definition and range helps you quickly eliminate wrong options.

Positive correlation amplifies risk when assets move together.
Negative correlation provides a natural hedge.

⚠️Common Mistake

Do not confuse correlation with causation. Two assets may move together because of a common market factor, not because one causes the other. The exam tests the statistical meaning, not the economic cause.

Pearson Correlation Coefficient

∑Formula: Pearson Correlation Coefficient (ρ)

\rho_{XY}=\frac{\sum_{i=1}^{n}(X_i-\bar X)(Y_i-\bar Y)}{\sqrt{\sum_{i=1}^{n}(X_i-\bar X)^2\;\sum_{i=1}^{n}(Y_i-\bar Y)^2}}

Where:

X_i= Return of asset X in period i

Y_i= Return of asset Y in period i

\bar X= Mean return of asset X over n periods

\bar Y= Mean return of asset Y over n periods

n= Number of observation periods

\rho_{XY}= Pearson correlation coefficient between X and Y

Worked Example

Given returns for 3 months: X: 5%, 10%, 15% Y: 4%, 9%, 14% Step 1: \bar X = (5+10+15)/3 = 10, \bar Y = (4+9+14)/3 = 9 Step 2: Numerator = (5-10)(4-9)+(10-10)(9-9)+(15-10)(14-9) = 25+0+25 = 50 Step 3: Denominator = sqrt[( (-5)^2+0^2+5^2 ) * ( (-5)^2+0^2+5^2 )] = sqrt[50*50] = 50 Step 4: \rho_{XY}=50/50 = 1 Verification: 50/50 = 1.

To compute the coefficient, collect historical return data for the two asset classes over the same time horizon – usually monthly or weekly returns for the past 12‑36 months. Align the data, calculate the mean of each series, and then apply the formula step‑by‑step.

The sign of the result is critical. A positive value means the assets tend to move in the same direction, while a negative value means they move opposite. Zero (or a value close to zero) indicates that the returns are largely independent.

For the exam, remember the three interpretation bands: >0.7 strong positive, 0.3‑0.7 moderate, <0.3 weak/low. This quick reference helps you answer multiple‑choice questions without lengthy calculations.

Correlation Across Major Asset Classes

In the Indian market, equities often exhibit a moderate positive correlation with corporate bonds (≈0.3‑0.5) because both are influenced by macro‑economic growth. Government securities tend to have lower correlation with equities (≈0.2‑0.4) due to their safe‑haven nature.

Commodities such as gold usually show a weak or slightly negative correlation with equities (‑0.1 to 0.2) especially during periods of market stress, making gold a useful diversifier. Real‑estate returns can be loosely linked to equities (≈0.3) but are more driven by supply‑demand dynamics and interest rates.

Understanding these typical ranges enables a distributor to construct a portfolio that balances risk and return, and to answer exam items that ask which asset pair offers the greatest diversification benefit.

Correlation Interpretation Bands Used in NISM Exams

Correlation Value	Interpretation	Typical Asset‑Class Example (India)
> 0.70	Strong positive – assets move together	Large‑cap equities vs. sector index
0.30 – 0.70	Moderate – some common drivers	Equities vs. corporate bonds
< 0.30	Low/weak – good diversification	Equities vs. gold

Sample Correlation Values Between Common Indian Asset Classes

Impact of Correlation on Portfolio Diversification

Diversification works because the total portfolio risk is not a simple weighted average of individual risks; correlation determines how risks combine. When two assets have low or negative correlation, the portfolio’s overall volatility is reduced compared to holding each asset separately.

The standard portfolio‑variance formula incorporates correlation explicitly. By adjusting asset weights and selecting assets with favorable correlation, a distributor can design a PMS that meets the client’s risk‑tolerance profile while aiming for target returns.

Exam questions often present weights, individual standard deviations, and a correlation coefficient, then ask for the portfolio’s standard deviation. Knowing the formula and the correct order of operations is essential.

∑Formula: Two‑Asset Portfolio Variance

\sigma_p^{2}=w_1^{2}\sigma_1^{2}+w_2^{2}\sigma_2^{2}+2w_1w_2\sigma_1\sigma_2\rho_{12}

Where:

w_1= Weight of asset 1 in the portfolio (decimal)

w_2= Weight of asset 2 in the portfolio (decimal)

σ_1= Standard deviation of asset 1 (decimal, e.g., 0.12 for 12%)

σ_2= Standard deviation of asset 2 (decimal)

ρ_{12}= Pearson correlation coefficient between asset 1 and asset 2

σ_p^{2}= Portfolio variance (decimal squared)

Worked Example

Given: w1 = 0.60, σ1 = 12% (0.12) w2 = 0.40, σ2 = 6% (0.06) ρ12 = 0.20 Step 1: w1^2 σ1^2 = 0.60^2 × 0.12^2 = 0.36 × 0.0144 = 0.005184 Step 2: w2^2 σ2^2 = 0.40^2 × 0.06^2 = 0.16 × 0.0036 = 0.000576 Step 3: 2 w1 w2 σ1 σ2 ρ12 = 2 × 0.60 × 0.40 × 0.12 × 0.06 × 0.20 = 0.0006912 Step 4: σ_p^2 = 0.005184 + 0.000576 + 0.0006912 = 0.0064512 Step 5: Portfolio SD σ_p = √0.0064512 = 0.0803 → 8.03% p.a. Verification: √0.0064512 = 0.0803 (rounded to 4 d.p.).

Example: NISM‑Style Portfolio Construction Scenario

Scenario

An Indian HNI wants a balanced PMS with 60% in large‑cap equities and 40% in high‑quality corporate bonds. Historical data shows equity SD = 14% p.a., bond SD = 5% p.a., and a correlation of 0.25 between them.

Solution

Using the two‑asset variance formula: w1 = 0.60, σ1 = 0.14; w2 = 0.40, σ2 = 0.05; ρ = 0.25. Compute each term: w1^2σ1^2 = 0.36 × 0.0196 = 0.007056; w2^2σ2^2 = 0.16 × 0.0025 = 0.000400; cross term = 2 × 0.60 × 0.40 × 0.14 × 0.05 × 0.25 = 0.00168. Add: 0.007056 + 0.000400 + 0.00168 = 0.009136. Portfolio SD = √0.009136 ≈ 0.0956 → 9.56% p.a. The diversification benefit reduces the portfolio risk from a simple weighted average (≈10.4%) to 9.56%. Thus the distributor can justify the allocation as meeting the client’s moderate‑risk profile.

Conclusion

The example demonstrates how a modest positive correlation still yields risk reduction. Remember to square the weights and use the correlation sign exactly as shown.

ℹ️Exam Trap

A frequent error is to omit the factor 2 in the cross‑term of the variance formula or to use the correlation value without converting percentages to decimals. Both mistakes give a higher variance than the correct answer.

Using Correlation in PMS Recommendations

When a distributor prepares a PMS proposal, they first calculate the historical correlation matrix for the client’s shortlisted asset classes. The matrix highlights pairs that offer the best diversification benefit.

Based on the matrix, the distributor can suggest weight adjustments – for example, increasing the allocation to an asset with low correlation to the dominant holding. SEBI’s risk‑management guidelines encourage such quantitative justification.

In the exam, you may be asked to identify which re‑balancing action will most effectively lower portfolio variance. Choose the asset pair with the lowest correlation and the highest individual volatility for the greatest impact.

Key Limitations and Considerations

Correlation is a historical, linear measure. It assumes a stable relationship over the chosen time window, which may break during market stress (e.g., equity‑bond correlations can rise in a crisis). Hence, distributors should monitor correlation trends regularly.

Non‑linear relationships, such as those captured by copulas, are beyond the NISM syllabus but are important for advanced risk management. For the exam, stick to the Pearson coefficient unless the question explicitly mentions another method.

Finally, the length of the observation period matters. Short windows can produce volatile correlation estimates, while very long windows may dilute recent market dynamics. SEBI recommends a minimum of 12 months for PMS risk assessments.

ℹ️Memory Aid

Remember ‘R for Relationship’: Correlation (R) tells you how two returns relate, not why they relate.

Regulatory Perspective

SEBI’s PMS regulations (Regulation 2(2) of the PMS Guidelines) require distributors to maintain a risk‑management framework that includes correlation analysis. The guidelines state that the risk‑monitoring system must track the correlation matrix of the portfolio’s constituent asset classes on a periodic basis.

Failure to incorporate correlation can lead to non‑compliance findings during SEBI inspections. Therefore, a distributor must be able to explain how correlation inputs affect the portfolio’s VaR or stress‑test outcomes.

Exam questions may reference SEBI’s emphasis on correlation, asking which statement best reflects the regulator’s intent – typically the answer highlights “monitoring of correlation among asset classes for risk mitigation.”

Rolling 6‑Month Correlation Between Equities and Gold (2023‑2024)

⭐Exam Takeaways

Correlation quantifies the linear relationship between two asset‑class returns and ranges from –1 to +1.
Use the Pearson formula; convert percentages to decimals before substitution.
Interpretation bands: >0.7 strong, 0.3‑0.7 moderate, <0.3 weak – useful for quick exam decisions.
Portfolio variance formula incorporates correlation; remember the factor 2 in the cross‑term.
Low or negative correlation between assets reduces overall portfolio risk, a core principle for PMS diversification.
SEBI mandates periodic monitoring of the correlation matrix as part of a PMS risk‑management framework.
Common exam traps: forgetting to square weights, using percentages instead of decimals, or mixing up correlation with causation.

Practice Questions

8 questions on Understanding Correlation Across Asset Classes

What is the range of values that a correlation coefficient can take?

Which statement correctly distinguishes correlation from causation?

Given monthly returns X: 5%, 10%, 15% and Y: 4%, 9%, 14%, what is the Pearson correlation coefficient between X and Y?

According to the NISM interpretation bands, a correlation of 0.45 is classified as:

An asset pair has w1=0.55, σ1=10% (0.10), w2=0.45, σ2=8% (0.08) and ρ=0.25. What is the portfolio standard deviation?

A distributor wants to lower portfolio variance most effectively. The current portfolio is 70% equities (σ=14%) and 30% corporate bonds (σ=5%) with ρ=0.25. Which re‑balancing action provides the greatest variance reduction?

Based on typical Indian market ranges, which asset‑class pair exhibits the weakest correlation?

Which SEBI regulation explicitly requires PMS distributors to maintain a risk‑management framework that includes periodic monitoring of the correlation matrix?

←Importance of Asset Allocation Decision Steps in Portfolio Management Process→