R help - Apr 2004 - GepR - bayesian modelling - please help!

Hi,

I am coming back with a question about GeoR hopping somebody 
can help.

I am doing a bayesian kriging and the prediction coordinates are 
shifted 0.1 m from the data coordinates.

I am using the following command and i am getting the following 
error:

zn.bayes <- krige.bayes(zn.gdata, loc = loc1, model = 
model.control(cov.model = "exponential", lambda = 0), 
prior = prior.control(phi.prior ="exponential", phi = 
89.1948), output=output.control(n.predictive=2, 
mean.var = TRUE, quantile = c(0.025, 0.25, 0.5, 0.75, 
0.975), threshold = c(300)))

Error: Error in cond.sim(env.loc = base.env, env.iter = iter.env, 
loc.coincide = get("loc.coincide",  : 
        chol: matrix not pos def, diag[13]= -3.487549e-019

What i am doing wrong? The location loc1 is a data frame with the 
shifted coord. values. Any suggestion is very well appreciated.

Thanks,

Monica

Monica

Could you please send me the
zn.gdata and loc1 objects and I will investigate this

P.J.

On Fri, 23 Apr 2004, Monica Palaseanu-Lovejoy wrote:
> Hi,
>
> I am coming back with a question about GeoR hopping somebody
> can help.
>
> I am doing a bayesian kriging and the prediction coordinates are
> shifted 0.1 m from the data coordinates.
>
> I am using the following command and i am getting the following
> error:
>
> zn.bayes <- krige.bayes(zn.gdata, loc = loc1, model >
model.control(cov.model = "exponential", lambda = 0),
> prior = prior.control(phi.prior ="exponential", phi >
89.1948), output=output.control(n.predictive=2,
> mean.var = TRUE, quantile = c(0.025, 0.25, 0.5, 0.75,
> 0.975), threshold = c(300)))
>
> Error: Error in cond.sim(env.loc = base.env, env.iter = iter.env,
> loc.coincide = get("loc.coincide",  :
>         chol: matrix not pos def, diag[13]= -3.487549e-019
>
> What i am doing wrong? The location loc1 is a data frame with the
> shifted coord. values. Any suggestion is very well appreciated.
>
> Thanks,
>
> Monica
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://www.stat.math.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
>
>
Paulo Justiniano Ribeiro Jr
Departamento de Estat??stica
Universidade Federal do Paran??
Caixa Postal 19.081
CEP 81.531-990
Curitiba, PR  -  Brasil
Tel: (+55) 41 361 3471
Fax: (+55) 41 361 3141
e-mail: pj at est.ufpr.br
http://www.est.ufpr.br/~paulojus

Hi again,

I was checking again everything after i sent you my data. It seems 
that the problem is the following. My prediction locations are 
different in number than my data locations ..... i forgot that when i 
built the geodata object few points had more than one value so i 
used the "first" option and my data number decreased. When i 
used the right number of locations shifted by 0.1 m the krige. 
bayes did work this time.

Now i have to figure out how much memory i do need to run the 
same function on a grid with about 15000 grid cells. If you have any 
info about that i will be really very grateful. 

Also i was searching for the "Ribeiro, P.J. Jr. and Diggle, P.J. 
(1999) _Bayesian inference in Gaussian model-based 
geostatistics_. Tech. Report ST-99-08, Dept Maths and Stats, 
Lancaster University.Available at: 
 <URL: http://www.maths.lancs.ac.uk/~ribeiro/publications.html>" 

The link is invalidated now.

Thank you so much for all your help and time,

Monica