Timeline for Calculation of standard variance $s^2$
Current License: CC BY-SA 3.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 22, 2014 at 17:24 | comment | added | jbowman | @whuber - I wish I could give you a lot more than 1 +1 for that "blatantly wrong" comment! | |
| Apr 21, 2014 at 18:47 | comment | added | whuber♦ | I'm sorry, it's not: please read @jbowman's comment. | |
| Apr 21, 2014 at 18:36 | comment | added | 1190 | As a result, my solution is true, Isnt it? @whuber | |
| Apr 21, 2014 at 17:48 | comment | added | whuber♦ | The answer is blatantly wrong on three counts. (1) Even under simplistic assumptions about the targeting errors (namely, that they are approximately Normal and do not favor any particular direction over any other), the degrees of freedom will not be $6$! They will be either $14$ or $21$ depending on whether errors are computed in two or three dimensions. (2) Even with $6$ df the rejection region would not extend all the way to $164$--it only goes to $12.6$. (3) Worse, this is the wrong rejection region! "Strong evidence" would be a significantly small (not large) value. Throw this book away! | |
| Apr 21, 2014 at 17:29 | comment | added | jbowman | You should be calculating $s^2$ around a "mean" of zero, not of 4.57. 4.57 is the average distance from the target; for an accuracy of this type, you want to know the standard deviation of where the missile hits around the target, not the standard deviation around the average distance from the target. | |
| Apr 21, 2014 at 16:35 | history | asked | 1190 | CC BY-SA 3.0 |