1)    What is 'best'?  This is not a rhetorical question.  Different
interpolation methods do different things, and some of those things
are more desireable than others, for a given situation.  Sorry if that
makes life more confusing :)

2)    Perhaps the simplest method is also the best.
(2a)    'best' may mean, the deviation between what you interpolate
and what you would actually detect, if you dug a 5th hole.  you can
estimate this by doing the interpolation on 3 of the holes, and see
how well it predicts the 4th.  If you dare :)
(2b)    Ask, what is the precision of the measurmeents at each
borehole.  No interpolation method should be asked to do better than
that.
(2c)    Any interpolation _assumes_ uniform, or 'slowly changing'
properties between holes.  (slowly changing = small 1st & 2nd
derivative)  Do you have reason to believe this assumption is false?
If so, then you must take this belief into account in your
interpolation scheme, whatever you choose.

If false, and you believe properties change slowly, then what are you
waiting for?  a simple method will do better, and get you close to
home sooner.

any help?

Cheers,
Jay

Ashraf A Ahmed wrote:

> Dear Sir,
>
> I have a statistical problem and would appreciate your guidance.
>
> I am doing my research in Civil and Environmental Engineering.
> I have a field site 155 ft length and 60 ft wide, where there are 4
> boreholes. I know the soil properties only at these boreholes. I
> need to generate these properties for the whole site.
> I have read that many interpolation options like Kriging, natural
> neighbor or inverse distance weighted can do that.
> But I don't know which method of those mentioned above gives me
> the best results. I would appreciate very much your hint in that.
> Also, is there any program which could be downloaded free from
> your website or any other website that can do that?.
>
> Thanks and hope to hear from you.
>
> ======================================
>
> Ashraf Ahmed
> Department of Civil Engineering
> The University of Manchester
> Oxford Road
> Manchester M13 9PL
> UK
>
> Phone: +44(0)161-275-4375
> Fax  : +44(0)161-275-4361
> .
> .
> =================================================================
> Instructions for joining and leaving this list, remarks about the
> problem of INAPPROPRIATE MESSAGES, and archives are available at:
> .                  http://jse.stat.ncsu.edu/                    .
> =================================================================

--
Jay Warner
Principal Scientist
Warner Consulting, Inc.
4444 North Green Bay Road
Racine, WI 53404-1216
USA

Ph: (262) 634-9100
FAX: (262) 681-1133
email: [EMAIL PROTECTED]
web: http://www.a2q.com

The A2Q Method (tm) -- What do you want to improve today?




.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to