NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
Re: That darned old cocked hat
From: Peter Fogg
Date: 2010 Dec 10, 09:16 +1100
From: Peter Fogg
Date: 2010 Dec 10, 09:16 +1100
George Huxtable wrote:
Well its not just me, anyone can improve on averaged values via comparison with the calculated slope. However, as I have recently and once again pointed out, this will only work if the random errors don't cancel each other out, as they effectively did in Antoine's recent example of a very good round of sights. In this special case averaging will work just as well. This is the underlying assumption that averaging relies upon. Of course, random errors may, or may not cancel themselves out so perfectly...
Use of slope still offers the advantage of seeing those sights in picture form and thus being offered the opportunity to evaluate them quickly by sight - something much more difficult to achieve with a row of numbers.
Your carefully contrived "artificially-generated observations" are irrelevant to this, George.
He has claimed that he can get closer to the
truth than can a statistical least-squares fit, or simply averaging the
observations for time and altitude.
Well its not just me, anyone can improve on averaged values via comparison with the calculated slope. However, as I have recently and once again pointed out, this will only work if the random errors don't cancel each other out, as they effectively did in Antoine's recent example of a very good round of sights. In this special case averaging will work just as well. This is the underlying assumption that averaging relies upon. Of course, random errors may, or may not cancel themselves out so perfectly...
Use of slope still offers the advantage of seeing those sights in picture form and thus being offered the opportunity to evaluate them quickly by sight - something much more difficult to achieve with a row of numbers.
Your carefully contrived "artificially-generated observations" are irrelevant to this, George.
