- you could modify the suggested approach by using a generalization of the

Poisson, the neg binomial assumption you mention. most stat software

allows negative binomial regression. in this case, the variance component

of the Chi-squared resids may be better approximated (than under the

Poisson assumption). as an aside, you may have a zillion zeroes with your

fisheries data. such data may be handled moderately well by the neg bin

assumption you mention. however, they may better be handled under the

assumption that some portion of the zeroes are structural (ie *can't*

generate a positive count) rather than stochastic. I haven't seen spatial

corr assessed under these assumptions in the published lit. regardless,

such "zero inflated" models are often considerably more complicated and may

not suit your purposes. brian

****************************************************************

Brian Gray

USGS Upper Midwest Environmental Sciences Center

2630 Fanta Reed Road, La Crosse, WI 54602

608-783-7550 ext 19 - Onalaska campus or

608-781-6234 - La Crosse campus

fax 608-783-8058

brgray@...

*****************************************************************

|---------+---------------------------->

| | "Edzer J. |

| | Pebesma" |

| | <e.pebesma@geog.u|

| | u.nl> |

| | Sent by: |

| | ai-geostats-list@|

| | unil.ch |

| | |

| | |

| | 11/28/2003 09:44 |

| | AM |

| | Please respond to|

| | "Edzer J. |

| | Pebesma" |

| | |

|---------+---------------------------->>--------------------------------------------------------------------------------------------------------------|

| |

| To: Marcelo Alexandre Bruno <marcelo2lei@...> |

| cc: ai-geostats@... |

| Subject: Re: AI-GEOSTATS: About gstat and binomial negative family data |>--------------------------------------------------------------------------------------------------------------|

Marcelo Alexandre Bruno wrote:

>Whats wrong? Someone could help me?

I don't know if something is wrong. Maybe your data don't exhibit

>

much spatial correlation, maybe they are so skew that without

transformation you just don't see any in sample variograms.

>The family of distribution of M. stehmanni is binomial

Not without modifying the source code. You could, as a first shot, try

>negative, is possible define these prior to variogram

>and then result better variograms?

>

to look at Pearson residual variograms assuming a Poisson distribution.

This can be done with gstat (be it a little forceful); look at the variance

argument to the gstat function, define beta and the covariates such that

the trend value is set for each observation. (If the trend is constant,

this

whole action is useless, given where you are now).

--

Edzer

--

* To post a message to the list, send it to ai-geostats@...

* As a general service to the users, please remember to post a summary of

any useful responses to your questions.

* To unsubscribe, send an email to majordomo@... with no subject and

"unsubscribe ai-geostats" followed by "end" on the next line in the message

body. DO NOT SEND Subscribe/Unsubscribe requests to the list

* Support to the list is provided at http://www.ai-geostats.org

--

* To post a message to the list, send it to ai-geostats@...

* As a general service to the users, please remember to post a summary of any useful responses to your questions.

* To unsubscribe, send an email to majordomo@... with no subject and "unsubscribe ai-geostats" followed by "end" on the next line in the message body. DO NOT SEND Subscribe/Unsubscribe requests to the list

* Support to the list is provided at http://www.ai-geostats.org - I know of a paper where people split up the process in begin zero or

positive (binomial), and the value of the process given that it is

positive (Poisson). In fact you're working with a composite pdf, two spatial

processes that have to be merged later on. The idea is attractive,

but not very easy. If you want the title of the paper, email me.

--

Edzer

Brian R Gray wrote:

>you could modify the suggested approach by using a generalization of the

--

>Poisson, the neg binomial assumption you mention. most stat software

>allows negative binomial regression. in this case, the variance component

>of the Chi-squared resids may be better approximated (than under the

>Poisson assumption). as an aside, you may have a zillion zeroes with your

>fisheries data. such data may be handled moderately well by the neg bin

>assumption you mention. however, they may better be handled under the

>assumption that some portion of the zeroes are structural (ie *can't*

>generate a positive count) rather than stochastic. I haven't seen spatial

>corr assessed under these assumptions in the published lit. regardless,

>such "zero inflated" models are often considerably more complicated and may

>not suit your purposes. brian

>

>****************************************************************

>

>

* To post a message to the list, send it to ai-geostats@...

* As a general service to the users, please remember to post a summary of any useful responses to your questions.

* To unsubscribe, send an email to majordomo@... with no subject and "unsubscribe ai-geostats" followed by "end" on the next line in the message body. DO NOT SEND Subscribe/Unsubscribe requests to the list

* Support to the list is provided at http://www.ai-geostats.org - sounds like you are describing a two-part or hurdle model. a possibly more

attractive but complex approach (zero-inflated count distributions)

postulates two sources of zeroes: structural and stochastic. this doesn't

require working with a zero-truncated count distribution. the downside is

that the process defining how zeroes are separated is latent. brian

****************************************************************

Brian Gray, Ph.D.

USGS Upper Midwest Environmental Sciences Center

2630 Fanta Reed Road, La Crosse, WI 54602

608-783-7550 ext 19 - Onalaska campus or

608-781-6234 - La Crosse campus

fax 608-783-8058

brgray@...

*****************************************************************

|---------+---------------------------->

| | "Edzer J. |

| | Pebesma" |

| | <e.pebesma@geog.u|

| | u.nl> |

| | |

| | 11/29/2003 06:35 |

| | AM |

| | |

|---------+---------------------------->>--------------------------------------------------------------------------------------------------------------------------------------------------|

| |

| To: Brian R Gray <brgray@...> |

| cc: ai-geostats@..., Marcelo Alexandre Bruno <marcelo2lei@...> |

| Subject: Re: AI-GEOSTATS: About gstat and binomial negative family data |>--------------------------------------------------------------------------------------------------------------------------------------------------|

I know of a paper where people split up the process in begin zero or

positive (binomial), and the value of the process given that it is

positive (Poisson). In fact you're working with a composite pdf, two

spatial

processes that have to be merged later on. The idea is attractive,

but not very easy. If you want the title of the paper, email me.

--

Edzer

Brian R Gray wrote:

>you could modify the suggested approach by using a generalization of the

may

>Poisson, the neg binomial assumption you mention. most stat software

>allows negative binomial regression. in this case, the variance component

>of the Chi-squared resids may be better approximated (than under the

>Poisson assumption). as an aside, you may have a zillion zeroes with your

>fisheries data. such data may be handled moderately well by the neg bin

>assumption you mention. however, they may better be handled under the

>assumption that some portion of the zeroes are structural (ie *can't*

>generate a positive count) rather than stochastic. I haven't seen spatial

>corr assessed under these assumptions in the published lit. regardless,

>such "zero inflated" models are often considerably more complicated and

>not suit your purposes. brian

--

>

>****************************************************************

>

>

* To post a message to the list, send it to ai-geostats@...

* As a general service to the users, please remember to post a summary of any useful responses to your questions.

* To unsubscribe, send an email to majordomo@... with no subject and "unsubscribe ai-geostats" followed by "end" on the next line in the message body. DO NOT SEND Subscribe/Unsubscribe requests to the list

* Support to the list is provided at http://www.ai-geostats.org - If you can handle writing (or get someone to write) MCMC code/bayesian

work then take a look a this zero inflated poisson with spatial effects.

Zero-Inflated Models with Application to Spatial Count Data

D.K.Agarwal,A.E.Gelfand and S.Citron-Pousty, Environmental and

Ecological Statistics 2002, vol 9, pp 341-355

Zero inflated poisson comes awfully close to a negative binomial, and

makes more "biological" sense (i.e. there are a lot more places with

zero counts and there is a process to account for them).

Thanks,

Steve

Brian R Gray wrote:

>sounds like you are describing a two-part or hurdle model. a possibly more

--

>attractive but complex approach (zero-inflated count distributions)

>postulates two sources of zeroes: structural and stochastic. this doesn't

>require working with a zero-truncated count distribution. the downside is

>that the process defining how zeroes are separated is latent. brian

>

>****************************************************************

>Brian Gray, Ph.D.

>USGS Upper Midwest Environmental Sciences Center

>2630 Fanta Reed Road, La Crosse, WI 54602

>608-783-7550 ext 19 - Onalaska campus or

>608-781-6234 - La Crosse campus

>fax 608-783-8058

>brgray@...

>*****************************************************************

>

>

>|---------+---------------------------->

>| | "Edzer J. |

>| | Pebesma" |

>| | <e.pebesma@geog.u|

>| | u.nl> |

>| | |

>| | 11/29/2003 06:35 |

>| | AM |

>| | |

>|---------+---------------------------->

> >--------------------------------------------------------------------------------------------------------------------------------------------------|

> | |

> | To: Brian R Gray <brgray@...> |

> | cc: ai-geostats@..., Marcelo Alexandre Bruno <marcelo2lei@...> |

> | Subject: Re: AI-GEOSTATS: About gstat and binomial negative family data |

> >--------------------------------------------------------------------------------------------------------------------------------------------------|

>

>

>

>

>I know of a paper where people split up the process in begin zero or

>positive (binomial), and the value of the process given that it is

>positive (Poisson). In fact you're working with a composite pdf, two

>spatial

>processes that have to be merged later on. The idea is attractive,

>but not very easy. If you want the title of the paper, email me.

>--

>Edzer

>

>Brian R Gray wrote:

>

>

>

>>you could modify the suggested approach by using a generalization of the

>>Poisson, the neg binomial assumption you mention. most stat software

>>allows negative binomial regression. in this case, the variance component

>>of the Chi-squared resids may be better approximated (than under the

>>Poisson assumption). as an aside, you may have a zillion zeroes with your

>>fisheries data. such data may be handled moderately well by the neg bin

>>assumption you mention. however, they may better be handled under the

>>assumption that some portion of the zeroes are structural (ie *can't*

>>generate a positive count) rather than stochastic. I haven't seen spatial

>>corr assessed under these assumptions in the published lit. regardless,

>>such "zero inflated" models are often considerably more complicated and

>>

>>

>may

>

>

>>not suit your purposes. brian

>>

>>****************************************************************

>>

>>

>>

>>

>

>

>

>

>

>

>

>

>--

>* To post a message to the list, send it to ai-geostats@...

>* As a general service to the users, please remember to post a summary of any useful responses to your questions.

>* To unsubscribe, send an email to majordomo@... with no subject and "unsubscribe ai-geostats" followed by "end" on the next line in the message body. DO NOT SEND Subscribe/Unsubscribe requests to the list

>* Support to the list is provided at http://www.ai-geostats.org

>

>

* To post a message to the list, send it to ai-geostats@...

* As a general service to the users, please remember to post a summary of any useful responses to your questions.

* To unsubscribe, send an email to majordomo@... with no subject and "unsubscribe ai-geostats" followed by "end" on the next line in the message body. DO NOT SEND Subscribe/Unsubscribe requests to the list

* Support to the list is provided at http://www.ai-geostats.org - Here are the answers I had for my question:

Thank you very much.

Marta

>>Dear list members,

Carme Hervada i Sala <carme.hervada@...> :

>>

>>

>>I would like to know if anyone has information or bibliography on

>>backtransformation of the variogram or the variogram model.

>>I have 2 ref. only (Armstrong and Guiblin et al. 1995).

>>Is this supose to give similar results to the log-normal kriging?

>>Could anyone point me bibliography for this, please....

>>

>>Thank you very much in advance,

>>any help would be appreciated,

>>Best wishes

>>Marta

>It depends on the transformation you do!

--

>See GSLIB- book (1992) for normal score transform

>see

>G. Mateu-Figueras et al: Normal in R+ vs lognormal in R., 2002 (iamg

>meeting, berlin september 2002)

* To post a message to the list, send it to ai-geostats@...

* As a general service to the users, please remember to post a summary of any useful responses to your questions.

* To unsubscribe, send an email to majordomo@... with no subject and "unsubscribe ai-geostats" followed by "end" on the next line in the message body. DO NOT SEND Subscribe/Unsubscribe requests to the list

* Support to the list is provided at http://www.ai-geostats.org