|
|
| Re: Canarsie CBTC (100597) | |||
 |
|||
| Home > SubChat | |||
|
[ Read Responses | Post a New Response | Return to the Index ] |
|
||
|
Re: Canarsie CBTC |
|
|
Posted by Stephen Bauman on Sun Jun 19 09:02:21 2005, in response to Re: Canarsie CBTC, posted by AlM on Sun Jun 19 06:43:34 2005. You are asking a fiendishly difficult question here.I do not think so. Perhaps, that is why Mr. Bayside has not yet responded. Let me restate something to make it a little clearer. I am asking for the accuracy, not the error. The accuracy is the maximum value of the error. Let me state it as a probability problem. X is a vector-valued random variable with expected value (0,0,0).... I think you are reading too much into the question I posed. But it's not all in one dimension, which makes the problem very difficult. I have taught basic college level probability but don't even want to think about the integrals you'd have to do to calculate that answer. The usual way for dealing with vectors, is to assume that all the vector components are independent gaussian random variables. Any rotation or translation of the axes means that the vectors will have independent gaussian random variables, when expressed in the new coordinates. Similarly a translation to polar coordinates results the radius length being a rayleigh random variable and the angle being a uniform random variable with both being independent of one another. |
|
|
|