Which of the following assumptions underlie the 'square root of time' rule used for computing volatility estimates over different time horizons?
I. asset returns are independent and identically distributed (i.i.d.)
II. volatility is constant over time
III. no serial correlation in the forward projection of volatility
IV. negative serial correlations exist in the time series of returns
Correct Answer: D
Explanation
The square root of time rule can be used to convert, say a 1-day volatility to a 10-day volatility, by multiplying the known volatility number by the square root of time to get the volatility over a different time horizon.
However, there are key assumptions that underlie the application of this rule, and statements I to III correctly state those assumptions. If serial correlations (whether negative or positive) exist, then asset returns are not independent as they are affected by the prior day or prior period's returns, and we cannot use the square root of time rule. Therefore Choice 'd' is the correct answer.
In order to use the 'square root of time' rule, asset returns should be iid, volatility should stay constant (ie there should be no volatility clustering), and no serial correlations (ie the returns of one day should not be affected by the returns of the prior periods). Choice 'd' is the correct answer.