How does the Central Limit Theorem apply to portfolio return estimation, and what sample size is 'large enough'?

The Central Limit Theorem (CLT) states that for a population with mean μ and finite variance σ², the sampling distribution of the sample mean X̄ approaches a normal distribution as the sample size n increases, regardless of the shape of the original population:

X̄ ~ N(μ, σ²/n) as n → ∞

The standard error of the mean is: SE = σ / √n

How Large Is 'Large Enough'?

The conventional threshold for CFA Level I is n ≥ 30. However:

• If the underlying population is already normal, CLT applies for any n
• If the population is moderately skewed, n ≥ 30 is usually sufficient
• For highly skewed distributions (e.g., hedge fund returns, venture capital), you may need n ≥ 100 or more

Practical Example — Windcrest Emerging Markets Fund

The Windcrest Emerging Markets Fund has monthly returns that are significantly right-skewed (fat right tail from occasional large gains). Historical data shows:

• Population mean monthly return: μ = 0.95%
• Population standard deviation: σ = 6.2%

An analyst takes a random sample of 36 months to estimate the fund's mean monthly return.

Using CLT, the sampling distribution of X̄ is approximately normal:

• Mean of X̄ = μ = 0.95%
• Standard error = σ / √n = 6.2% / √36 = 6.2% / 6 = 1.033%

What is the probability that the sample mean exceeds 2.0%?

z = (2.0% - 0.95%) / 1.033% = 1.05 / 1.033 = 1.016

P(X̄ > 2.0%) = P(z > 1.016) ≈ 15.5%

Even though the individual monthly returns are skewed, the CLT lets us use the normal distribution for the sample mean because n = 36 ≥ 30.

Why CLT Matters in Finance:

1. Confidence intervals for mean returns rely on normality of X̄, not normality of individual returns
2. Hypothesis testing about population means uses the standard error, which shrinks with larger n
3. Risk estimation — when aggregating many small, independent risks (insurance, credit portfolios), CLT explains why the aggregate loss distribution becomes bell-shaped

Exam Tip: If a question states n < 30 and the population is non-normal, you cannot invoke CLT. If n ≥ 30, you can use it even if the population is skewed. This is one of the most frequently tested distinctions in CFA Level I quantitative methods.

Practice more sampling and estimation problems in our CFA Level I question bank.