Timeline for Why is asset volatility easier to estimate than the asset mean if it contains the mean?
Current License: CC BY-SA 4.0
13 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Nov 12, 2020 at 13:47 | comment | added | fes | @RichardHardy Sure it is a bit unclear what is meant by relative precision. I think the notion that means are harder to estimate than variances arose in the literature that followed Markowitz’s classic paper and tried to study the effects of estimation uncertainty on mean-variance analysis. What these people meant by this statement is that estimation uncertainty about mean is the reason making mean-variance analysis problematic so for their purposes variance is better estimated. But I think you can see the point by simply looking at confidence intervals. | |
| Nov 12, 2020 at 11:02 | comment | added | Richard Hardy | Right. The way you construct confidence intervals is based on a very concrete mapping between the parameters of the distribution and the interval. Then we essentially can discuss the interval and the parameters interchangeably. But his main point clarified in the edit seems to be that the original problem is an apples-to-oranges comparison. That is an interesting point and a troubling one if we accept it, don't you think so? | |
| Nov 12, 2020 at 9:59 | comment | added | fes | @RichardHardy I think he could do more to demonstrate how the answer depends on the parameters of the distribution. The argument that looking at confidence intervals is pointless because there is a continuum of them is unorthodox and most statisticians would disagree. It would be similar to saying that 1% value at risk is pointless because you as well look at 5% value at risk. | |
| Nov 12, 2020 at 7:42 | comment | added | Richard Hardy | What do you think about the answer of Dave Harris? I have also posted a related question. | |
| Nov 9, 2020 at 15:29 | history | edited | fes | CC BY-SA 4.0 |
deleted 117 characters in body
|
| Nov 9, 2020 at 15:22 | history | edited | fes | CC BY-SA 4.0 |
deleted 11 characters in body
|
| Nov 9, 2020 at 15:13 | history | edited | fes | CC BY-SA 4.0 |
added 433 characters in body
|
| Nov 9, 2020 at 9:51 | comment | added | fes | @RichardHardy Another typo; corrected. Thanks! | |
| Nov 9, 2020 at 9:50 | history | edited | fes | CC BY-SA 4.0 |
added 15 characters in body
|
| Nov 9, 2020 at 9:40 | comment | added | fes | @RichardHardy You have the square root because it is the confidence interval for standard deviation. The latter was just a typo. I corrected it. Thanks! | |
| Nov 9, 2020 at 9:31 | history | edited | fes | CC BY-SA 4.0 |
added 144 characters in body
|
| Nov 8, 2020 at 20:36 | history | edited | fes | CC BY-SA 4.0 |
added 455 characters in body
|
| Nov 8, 2020 at 20:17 | history | answered | fes | CC BY-SA 4.0 |