00:01
So let's talk about standard deviation.
00:03
Standard deviation is a measure of how much the observations, the xi, vary from their mean.
00:11
We are thinking about how much the observations, in this case the portfolio returns, are different from the mean.
00:17
And we want to get a measure of how much noise is in the data.
00:20
So the first thing we want to do is add these up, figure out how far are these deviations from the mean are.
00:25
Are these things, are the observations close to the mean? or are they scattered a long way from the mean? but we also wouldn't then have that the down noise cancels out the up noise, and we don't want that.
00:37
So we square these terms to make sure they're positive, and then we divide.
00:42
We divide by n for a population, and we usually divide by n minus one for a sample.
00:53
And why we get into this, quote, finite sample correction is way too much for this video.
00:59
But suffice to say that the first step is that we need to get the average...