Which of the following assumptions underlie the 'square root of time' rule used for computing VaR estimates over different time horizons?
I. the portfolio is static from day to day
II. asset returns are independent and identically distributed (i.i.d.)
III. volatility is constant over time
IV. no serial correlation in the forward projection of volatility
V. negative serial correlations exist in the time series of returns
VI. returns data display volatility clustering
Correct Answer: C
Explanation
The square root of time rule can be used to convert, say a 1-day VaR to a 10-day VaR, by multiplying the known number by the square root of time to get the VaR over a different time horizon. However, there are key assumptions that underlie the application of this rule, and statements I to IV correctly state those assumptions.
Statements V and VI are not correct, because the application of the square root of time rule requires the absence of serial correlations, and also the absence of volatility clustering (ie independence). Therefore Choice
'c' is the correct answer.
The square root of time rule is also applied to convert volatility or standard deviation for one period to the volatility for a different time period. Remember that VaR is just a multiple of volatility, and therefore the assumptions that apply to the square root of time rule for VaR also apply to the same rule when used in the context of volatilities or standard deviation.