Final answer:
The statement is true because the Risk Metrics approach assumes a normal distribution that can underestimate extreme market fluctuations, which are outside the scope of a normal distribution. The Student's t distribution gets closer to the normal distribution as the sample size increases, but when sample sizes are small, it provides a better approximation due to its heavier tails.
Step-by-step explanation:
A criticism of the Risk Metrics approach to estimating market risk is that you must use a normal distribution which may not accurate reflect the probability of extreme events. The statement is true. The Risk Metrics methodology often assumes a normal distribution of market returns, which is known to underestimate the likelihood of extreme market movements, also known as "fat tails". The central limit theorem supports the idea that with large sample sizes, the sampling distribution of the means turns into a normal distribution. However, this approximation does not always accommodate the occurrence of extreme values that are outside the scope of a normal distribution.
The Student's t distribution provides a more accurate representation for smaller sample sizes (<30) as it accounts for the additional uncertainty by providing heavier tails than the normal distribution. The true value of a population mean from repeated samples would be contained in approximately 90% of the confidence intervals calculated, according to the theoretical distribution. Though a normal distribution can be a good approximation under certain circumstances, its usage is limited when dealing with non-normal data distributions, especially when the sample size is inadequate for the central limit theorem to hold true.