NavList:

   

Reply

Dear George,
I am surprised that that one rejected distance in
my first lunar triggered so much exchange.
I think this is based on some misunderstandings,
and maybe my statements in this exchange were not clear
enough.

Let me begin with two statements which I want to make
completely clear:

1. What does it exactly mean: "rejected an observation".
I did not erase it from my log. When necessary I reduced it
too and the result is kept in my log.
I rejected it IN COMPUTING THE AVERAGE.
The purpose of computing and reducing the average is to
obtain the "final result", let's say, my chronometer correction.
I want my chronometer correction to be as close to the TRUTH
as possible. Could I make any use of this "rejected" distance?
Yes, I could. When I would judge about the likely error
of my observations, I WOULD take it into account.
But to compute my chronometer correction I decided to reject it.
The later reduction of rejected distance PROVED that
for this particular series I was right:
rejecting this observation improved the average.

2. In my explanation why I rejected it, I did not mean to
formulate any general rules for rejection.
I rejected it using my common
sense.
And then, when asked, tried to analyse and explain how my
common sense worked in this particular case.
If I am asked to state a general rule for rejection, the
only rule would be:
USE YOUR COMMON SENSE.

Now let me comment some points of your message.
On Wed, 27 Oct 2004, George Huxtable wrote:

> I ask him if he can quote any authority in support
> of such a dogmatic attitude.

The only authority is my common sense.
I repeat this is not a dogmatic attitude:-)
The statement was nor a "rule", not a "dogma", but
an attempt to explain how my common sense worked in this
particular case.

> Pursuing this matter, I have asked whether the rejected point in his first
> lunar set would still have been rejected if the discrepancy had been an
> arc-minute less-

Such questions can be asked to infinity:-)
When my son was 4 years old I told him that the highest
mountain is Everest. His next question was: "What is the lowest mountain?"
I tried to explain him why this question has no reasonable
answer. Do you really want me to repeat this explanation
on this list?

> That shows a welcome flexibility creeping in, but the picture I get is that
> Alex is very reluctant to abandon his self-imposed rule.

I repeat: the only "self imposed rule" was:
use common sense. (This rule I will never abandon:-)
Again: I never wanted to claim that EVERY deviation from monotonicity
has to be rejected in EVERY set of data.
And sorry if my statements could be interpreted like this.
I was talking of THIS PARTICULAR series, nothing else.

A question to George:
Suppose you are in Sea, have no radio, and taking lunars
to compute your chronometer correction.
And you obtained exactly that series of lunar distances I had.
How would you proceed? Would you average all 6 or reject one?

> Finally, I asked Alex this question about his second set of lunars-.

> be interesting to know exactly what were those averaged values that were
> fed back into Frank Reed's program, and whether any rounding had taken
> place.

Those averaged values were given in my observation reports: In the first
report: AVERAGE GMT: 4:13:13 AVERAGE DIST: 23.54' In the second report:
AVERAGE GMT: 0:26:28
AVERAGE DIST: 70d41.0'

They were rounded to 1 sec of time and .1' of distance.
Which I always do, of course: if the almanac itself does it,
what is the point of messing with extra digits myself?

> So there's no
> discredit to Frank's program.

Who was talking on "discredit"???
Of course, his program also does rounding, how else can it be??!

Alex.

   

Reply