September 1st, 2004, 12:52 pm
Note: It sounds as though you are calculating your standard deviation using 1 month's worth of daily values? If this is the case then you are calculating a daily, not monthly standard deviation. The descriptor refers to the frequency of observations, not the time frame over which they were collected. If this is the case, then when you say that you wish to 'annualize' the standard deviation, do you really mean that you wish to calculate the standard deviation of annual returns?, or do you mean that you want the standard deviation for one year's worth of daily returns? The standard deviation for annual returns would theoretically be equal to the standard deviation of daily returns multiplied by the square root of 365 (the number of observations) (assuming that you are using log returns). The daily standard deviation for one year's worth of daily observations, however, would (theoretically) be the same as the daily standard deviation for one month's worth of observations, assuming your daily returns are i.i.d.If I have misunderstood, and you are calculating your standard deviation using monthly observations, then please disregard these thoughts.