NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
Re: Systematic error and its resolution
From: Peter Fogg
Date: 2007 Apr 7, 16:47 +1000
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to NavList@fer3.com
To , send email to NavList-@fer3.com
-~----------~----~----~----~------~----~------~--~---
From: Peter Fogg
Date: 2007 Apr 7, 16:47 +1000
George Huxtable contributed:
Yes. Just as "over simplistic" as assuming that there is only one chance in four that the position will lie within enclosing position lines, and for the same reason: in that case the "over simplistic" assumption relies on the notion that there is NO systematic error.
In practice a round of sights, particularly those involving random sights (by which I mean single sights to bodies that preclude an analysis of a succession of consecutive sights) could well involve unknown quantities of systematic AND random error.
So what effect will unknown quantities of systematic error have on the "over simplistic" one chance in four notion? I suspect they may stuff it up.
[For those unfamiliar with English idiom, this technical term indicates that this hitherto unconsidered variable may well, all by itself, render the hypothesis generally untenable.]
To whatever extent systematic error is present its effects on a triangle are clear: with azimuths less than 180 degrees, the corrected fix will tend to lie outside the triangle, while with >180 the corrected fix will tend to lie at the centre. Just where it has always been assumed to be. With good reason, perhaps.
Does the question come down to which effect is the greatest: random or systematic error?
I have been impressed (from a safe distance) by those who have devised computer simulations of random error to test the 'one in four' idea. Could these simulations be adapted to include different quantities of random AND systematic error, to see what the effects are of changing the quantities of each?
Can this be tested?
Peter's conclusions, such as that his "improved" fix will always lie inside
the triangle if the azimuths spread by more than 180 degrees, and, indeed
will be at its centre, are over simplistic, in that they apply only if it is
known that there is no random scatter ...
Yes. Just as "over simplistic" as assuming that there is only one chance in four that the position will lie within enclosing position lines, and for the same reason: in that case the "over simplistic" assumption relies on the notion that there is NO systematic error.
In practice a round of sights, particularly those involving random sights (by which I mean single sights to bodies that preclude an analysis of a succession of consecutive sights) could well involve unknown quantities of systematic AND random error.
So what effect will unknown quantities of systematic error have on the "over simplistic" one chance in four notion? I suspect they may stuff it up.
[For those unfamiliar with English idiom, this technical term indicates that this hitherto unconsidered variable may well, all by itself, render the hypothesis generally untenable.]
To whatever extent systematic error is present its effects on a triangle are clear: with azimuths less than 180 degrees, the corrected fix will tend to lie outside the triangle, while with >180 the corrected fix will tend to lie at the centre. Just where it has always been assumed to be. With good reason, perhaps.
Does the question come down to which effect is the greatest: random or systematic error?
I have been impressed (from a safe distance) by those who have devised computer simulations of random error to test the 'one in four' idea. Could these simulations be adapted to include different quantities of random AND systematic error, to see what the effects are of changing the quantities of each?
Can this be tested?
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to NavList@fer3.com
To , send email to NavList-@fer3.com
-~----------~----~----~----~------~----~------~--~---