I am working through a question and I have to annualize the residual error which is given as:
w = 1% quarterly
then the answer annualilizes this by doing 1 * SQRT (4) = Annualized w of 2…
I cant not for the life of me understand how this annulizes the residual error… anyone?
Variances are additive, so,
W1² + W2² + W3² + W4² = Wannual²
If all of the quarterly residuals are (assumed to be) the same (=W), then,
Wannual² = W² + W² + W² + W²
=4W²
So,
√(Wannual²) = √(4W²)
Wannual = 2W
OK I have a follow up question then. In the same question it gives the annual residual returns of 0.5 quarterly. And then they solve for the annual buy doing .5 +.5 +.5 +.5 = 2
Shouldnt it be 0.5 squared + 0.5 squared + 0.5 squared + 0.5 squared = 2 squared = SQRT 2 = 1?
Nope.
It’s akin to calculating total returns: technically they should be compounding, but they’re making it simpler by summing.