Re: [R-sig-Geo] variogram question

2008-08-22 Thread G. Allegri
[EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] En nombre de G. Allegri > Enviado el: 21 August 2008 00:09 > Para: Paul Hiemstra > CC: r-sig-geo@stat.math.ethz.ch > Asunto: Re: [R-sig-Geo] variogram question > >> You spoke of more sophisticated methods of automatically choosing be

Re: [R-sig-Geo] variogram question

2008-08-21 Thread Pilar Tugores Ferra
---Mensaje original- De: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] En nombre de G. Allegri Enviado el: 21 August 2008 00:09 Para: Paul Hiemstra CC: r-sig-geo@stat.math.ethz.ch Asunto: Re: [R-sig-Geo] variogram question > You spoke of more sophisticated methods of automatically cho

Re: [R-sig-Geo] variogram question

2008-08-20 Thread G. Allegri
> You spoke of more sophisticated methods of automatically choosing between > models, what kind of methods did you have in mind? Here's a list I gathered some time ago. Many of them are just buzzwords form me! ‧ Adjusted R-squared (Wherry 1931) ‧ Bootstrap (Efron 1979) ‧ Cross-validation (Stone

Re: [R-sig-Geo] variogram question

2008-08-20 Thread Paul Hiemstra
Hi, There will certainly be situations when the SS method will select an appropriate model. But there are also situations when it will not make a well thought over decision. For example if there is not a lot of data on short range variability the SS method will not be able to fit the model th

Re: [R-sig-Geo] variogram question

2008-08-20 Thread G. Allegri
> At this stage I do > this by computing the sums of squares between the model and the sample > variogram and choose the one with the smallest SS. This is a rather crude > way of selecting between the models. Thanks Paul for automap. I'm planning to try it in the next occasion. What are the major

Re: [R-sig-Geo] variogram question

2008-08-20 Thread Paul Hiemstra
Hi Wesley, Good to know that the package helped you. A note the choice between the different variogram models. At this stage I do this by computing the sums of squares between the model and the sample variogram and choose the one with the smallest SS. This is a rather crude way of selecting b

Re: [R-sig-Geo] variogram question

2008-08-19 Thread Wesley Roberts
Dear Paul and the rest of the users who replied to my question, Firstly many thanks for all your input, reading your emails this morning improved my mood exponentially. I have installed automap and am getting to know the program nicely, it is so easy to run it is almost unfair. I do have one q

Re: [R-sig-Geo] variogram question

2008-08-18 Thread Paul Hiemstra
...in addition, any feedback on the package would be more than welcome! Paul Edzer Pebesma schreef: In general: no, in special cases: yes. Fitting variograms involves non-linear regression for most models (Sph, Exp, Gau, ...) for the range parameter, so you need starting values. Given the in

Re: [R-sig-Geo] variogram question

2008-08-18 Thread Paul Hiemstra
Hi, The package Edzer was talking about, automap, can be downloaded from http://intamap.geo.uu.nl/~paul/Downloads.html. It makes a few practical, somewhat arbitrary assumptions about initial starting values. - initial nugget is lowest semivariance found in the sample variogram - initial sill

Re: [R-sig-Geo] variogram question

2008-08-18 Thread Paulo Justiniano Ribeiro Jr
A possible extension of Edzer thoughts on this is to consider a initial search for the non-linear range parameters. For instance, take a grid of (20 say) values for this parameters within a reasonable range for the given problem, e.g. between 0 and the maximum distance. Evaluate the objective func

Re: [R-sig-Geo] variogram question

2008-08-18 Thread Edzer Pebesma
In general: no, in special cases: yes. Fitting variograms involves non-linear regression for most models (Sph, Exp, Gau, ...) for the range parameter, so you need starting values. Given the initial range, linear regression is sufficient to find the nugget/sill component(s), as they are linear.

[R-sig-Geo] variogram question

2008-08-18 Thread Wesley Roberts
Dear r-sig-geo users, I am currently analyzing some Lidar data we have collected over our study area. I am interested in identifying the range of the semi-variogram as this value will determine the width of pseudo-flight lines I intend to use to sample the lidar data. Our point density is upwar