setting values for replicates

[monisola olaniyan]

Hi All,

 

I ran maxentwith 15 replicates and I got differnt values in each of the replicate. (e.g AUC of test and training data are not the same in all the 15 and I have different graphs and varaible contribution in all cases).

 

How do I know which one to pick among these options.

 

My vegetation variable which is the only categorical variable produced a different graph in all cases. It comes vertical unlike others in slopy shape.

 

Please am I missing something somewhere?

 

Thanks.  

[James Douglas]

You would want to use the average at the end. The average compiles the values for the replicates into a single output. So rather than selecting one of the replicates, choose the one that averages them all together.

--
You received this message because you are subscribed to the Google Groups "Maxent" group.
To post to this group, send email to max...@googlegroups.com.
To unsubscribe from this group, send email to maxent+un...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/maxent?hl=en.

[Martin Damus]

Don't forgot to look at the minimum and maximum projections. If they are very different then one interpretation is your locations are not well distributed across environmental space, and subsets of your locations result in very different projections.

Martin Damus

---

From: James Douglas <jhdou...@gmail.com>
To: max...@googlegroups.com
Sent: Thursday, August 16, 2012 12:12:56 PM
Subject: Re: setting values for replicates

[monisola olaniyan]

Thank you all for your contribution. I am new but contributions are really helping. Few more questions please:

 

 

There is an excel file for average value for each of my replicate- does that means I will sum all and find average? And, what about my prediction map, which one will I go for.

 

If I may get what minimum and maximum projections mean... is that 1st replicate and last replicate respectively?

 

I checked the maps and graphs and no significant difference between the 1st and last replicates.

Thanks so much.

[Florencia Sangermano]

Hi,

 

Maxent automatically calculates the average, max, min and median. You can find them in your output folder they will have the names:

Yourspecies_avg.asc

Yourspecies_min.asc

Yourspecies_max.asc

Yourspecies_median.asc

 

Minimum means that the resulting map shows for each pixel the minimum value across all runs.

Martin has a good observation, instead of looking at min and max values I usually look at the standard deviation (Yourspecies_stddev.asc) for the same purpose.

 

 

Also, maxent creates an html file with the summary of all runs (Yourspecies.html) – within this file and  under the heading “Pictures of the model” you will be able to see the average distribution and links to the min-max and median images

 

Flor

[monisola olaniyan]

Thank you Florence. I'm very grateful for the input.