NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
Re: Taking four stars for checking accuracy of fix - and "Cocked Hats"
From: Peter Fogg
Date: 2008 Aug 4, 15:57 +1000
From: Peter Fogg
Date: 2008 Aug 4, 15:57 +1000
Paul Hirose writes: > > Finding the "best fit" simply > requires taking a mean. No. Finding the best fit to a known slope involves an analysis of each sight and the resulting grouping relative to the slope, and a judgement of what weighting, ranging from absolute; the point on the line, to nil; the point effectively discarded, to give that data point, relative to the others. > What I get, for all four data sets, are > altitudes identical to the previous ones! (By "identical" I mean they > agree to the .01'; I didn't write down the values any more precisely.) > At least in this case, it seems that knowing the correct slope of the > line is no benefit. I remember Fred, some years ago now, showing us a lovely line of sights he'd made that matched almost perfectly the slope over a short period. In such a case then a regression analysis would have given the same result as a slope analysis. Indeed in such a case no analysis would have been necessary. Horses for courses. I block messages that contain attachments or HTML. I have dutifully changed this to Plain Text. Don't you tend to miss a lot of messages that may be inappropriately formatted? --~--~---------~--~----~------------~-------~--~----~ Navigation List archive: www.fer3.com/arc To post, email NavList@fer3.com To , email NavList-@fer3.com -~----------~----~----~----~------~----~------~--~---