Help needed

[Greg W]

Hi,
I'm trying to figure out how to creating an Excel spreadsheet based on the lambda file from a Maxent model.  I've based it on the Guidelines for computing MaxEnt model output values from a lambdas file PDF, by Peter D. Wilson that I found online.

I'm doing this as the model could possible be used by those who don't know how to use Maxent to run a set of data based on a previous model, or don't know how to use or have access to R (mainly for my advisor), and who are familiar with Excel.

I've been able to get a spreadsheet to accurately duplicate the background output from a system that is not using Theshold features, but when using a different model which includes threshold features, the spreadsheet results are way off.  Below is what the Guidlines PDF says regarding threshold features.

(f) Threshold features:
(263.5<OZ_AnnPrecip), 0.027969395068903, 0.0, 1.0
Recognised by the fact that the first part begins and ends in round brackets and reads like an
inequality expression (which is exactly what it represents). The value to the left of the less-than sign
is the threshold value. The second part is, as always, the fitted lambda-value, and the min and max
values represent the outcome of applying the threshold test. That is, 0 if true and 1 if false.
if x < threshold then fx = 0 otherwise fx = lambda

The major difference I'm seeing is that my threshold value is expressed differently, such as "(HYDRO_TYPE=0.0)" instead of "263.5<OZ_AnnPrecip)" as shown in the Guide.  I'm taking this to read as HYDRO_TYPE=0.0 instead of 0.0<HYDRO_TYPE, thinking that the values could be <,>,= and the Guide is showing just one possibility.

I've verified with a calculator that the spreadsheet calculations are correct, so I'm not sure what I am doing wrong.

I'm attaching  an example spreadsheet that contains the data and calculations that I'm referring to and shows both what the spreadsheet is calculating,and Maxent is returning.

I need someone to look over this and see what the issue is.   If someone could get me in contact with Peter Wilson who wrote the Guidelines PDF or anyone else who could help, that would be highly appreciated.  This is about 1/4 of my thesis, which is due to be defended at the end of this summer's semester.

Thanks,

Greg W.

[Jamie M. Kass]

Are you sure that HYDRO_TYPE is not a categorical variable? Just a thought. If it is, I think you need to specify this in the GUI.

[Greg W]

Hi Sir,
Yes, HYDRO_TYPE and REC_PARKS are both categorical.   It is specified as such as the instructions on the complete document, sorry, didn't include that here.

[ndimhypervol]

If that's the case, I am pretty sure they are handled by the categorical feature class, not threshold. That's why you see HYDRO_TYPE = 0, because it's giving you the beta value for the 0 category of hydro type.

Jamie Kass

PhD Candidate, CCNY

cell: 917-602-5787

--
You received this message because you are subscribed to a topic in the Google Groups "Maxent" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/maxent/vkZbNaBTWps/unsubscribe.
To unsubscribe from this group and all its topics, send an email to maxent+un...@googlegroups.com.
To post to this group, send email to max...@googlegroups.com.
Visit this group at https://groups.google.com/group/maxent.
For more options, visit https://groups.google.com/d/optout.

[Greg W]

Hi All,
Thanks for the replies.  Both led me to take another look and I've figured it out.  Still trying to make the macro work, but the spreadsheet attached now matches exactly.  I apparently misunderstood how the threshold feature is supposed to be calculated.

Thanks again,

Greg W.