NavList:

   

Reply

Alex:

I thought you said previously that we could consider the dip to be
constant over some period of time, like three hours. I agree that if
the dip is changing quickly so that for a given set of observations it
could not be considered reasonably constant, then there would be
problem. Same goes for clock error. But one could try different
experiments and look at the residuals from the LS analyses to help
gauge which experiments gave the best results (residuals should be
consistent with the probability distribution of the random errors in
the observations).

-- Richard

On 10-Apr-12, at 11:56 AM, Alexandre E Eremenko wrote:

>
> Richard,
>
>> As you say in your P.S., you HAVE to have a functional model
>> describing your data for parametric least squares. So your proposed
>> numerical example is incomplete.
>
> Of course. The example was given just to simply illustrate the
> statement in the PS.
> There is NO appropriate mathematical model for the anomalous dip.
> (The model should explain the change of the dip with time!)
>
> That's why statistics will not help with the dip.
>
> Alex.
>
>
>

-----------------------------------------------------------------------------
| Richard B. Langley                            E-mail:
lang@unb.ca         |
| Geodetic Research Laboratory                  Web: http://www.unb.ca/GGE/
  |
| Dept. of Geodesy and Geomatics Engineering    Phone:    +1 506
453-5142   |
| University of New Brunswick                   Fax:      +1 506
453-4943   |
| Fredericton, N.B., Canada  E3B
5A3                                        |
|        Fredericton?  Where's that?  See: http://
www.fredericton.ca/       |
-----------------------------------------------------------------------------

   

Reply