Skip to content
Get started

Does daily data pin down the drift?

An asset's log-returns are i.i.d. with annual volatility 20%20\%. You estimate its annual drift (expected annual log-return) from exactly 1010 years of price history by averaging the observed log-returns and annualising.

A colleague says switching from monthly data (120120 returns) to daily data (25202520 returns, 252252 trading days per year) will shrink the standard error by a factor of 21\sqrt{21}. What is the standard error of the annual drift estimate with daily data, in percent?

Show hints (2)+
  1. Write the sum of the log-returns over the whole period. What does it telescope to?
  2. Annualising multiplies both the mean and its error by 252.

Answer

Reveal answer →

6.32 (± 0.02)

Create a free account to see the full worked solution and track your progress.

Interview format: Hedge Fund Quant, Systematic & Data-Driven Trading

Related questions