GOF in U-CARE

questions concerning analysis/theory using programs M-SURGE, E-SURGE and U-CARE

GOF in U-CARE

Postby annaren » Tue Mar 11, 2008 7:28 am

I am using UCARE to test my GOF but I have a few queries which I can not find the answer to in either the User's manual or other papers using the software so I was wondering if you could help!

I have four groups:
0 1 0 1 - female narrow
0 1 1 0 - female wide
1 0 0 1 - male narrow
1 0 1 0 - male wide

First query is how to identify which group U-CARE is refering to. Does the group numbering that U-CARE uses correspond to the order that they are in the data, i.e the first block of capture histories are female narrow so would this be group one etc?

I want to test the general model phi(sex*width*time) p(sex*width*time)

I have ran the global test which is highly significant and also the individual tests.
TEST2.CL can not be calculated as trap occassions is <5.
TEST3.SR is significant for 3 of the groups
TEST3.SM is the same for all groups and is non-significant
TEST2.CT is significant for only one of the groups but U-CARE does not give results for group 4

Second query is why do I have no results for group 4 in TEST 2.CT?

Third query is why if I add up the individual quadratic chi2 values and also the df from each of the tests do they not equal that given in the global test?

Fifth query is if the global test shows both transience and I want to calulate the GOF of the transient model do I just add up the quadratic chi2 and df for each group for each test minus TEST3.SR?

Fifth query is if the global test shows both transience and trap dependence and I want to calulate the GOF of the transient and trap dependence model do I just add up the quadratic chi2 and df for each group for each test minus TEST3.SR and TEST 2.CT (therefore only TEST 3.SM in my case)?

Thank you so much for your help. I really apprecaite it.

Yours
Anna
annaren
 
Posts: 25
Joined: Tue Mar 04, 2008 12:00 pm
Location: University of Aberdeen

Postby CHOQUET » Fri Mar 14, 2008 5:32 am

Hello,

In reply to :

> I am using UCARE to test my GOF but I have a few queries which I can > not find the answer to in either the User's manual or other papers
> using the software so I was wondering if you could help!

> I have four groups:
> 0 1 0 1 - female narrow
> 0 1 1 0 - female wide
> 1 0 0 1 - male narrow
> 1 0 1 0 - male wide

> First query is how to identify which group U-CARE is refering to.
> Does the group numbering that U-CARE uses correspond to the order > that they are in the data, i.e the first block of capture histories are
> female narrow so would this be group one etc?

YES !!!!

> I want to test the general model phi(sex*width*time) p
> (sex*width*time)

> I have ran the global test which is highly significant and also
> the individual tests.
> TEST2.CL can not be calculated as trap occassions is <5.
> TEST3.SR is significant for 3 of the groups
> TEST3.SM is the same for all groups and is non-significant
> TEST2.CT is significant for only one of the groups but U-CARE
> does not give results for group 4

> Second query is why do I have no results for group 4 in TEST 2.CT?

It is not possible to guess without the data set. Certainly, there is a lack of data to compute the test. Look for the m-array.

> Third query is why if I add up the individual quadratic chi2 values and > also the df from each of the tests do they not equal that given in the
> global test?
> Fifth query is if the global test shows both transience and I want to
> calulate the GOF of the transient model do I just add up the quadratic > chi2 and df for each group for each test minus TEST3.SR?

Yes. It is the same procedure than for calculating the chat.

> Fifth query is if the global test shows both transience and trap
> dependence and I want to calulate the GOF of the transient and trap
> dependence model do I just add up the quadratic chi2 and df for each > group for each test minus TEST3.SR and TEST 2.CT (therefore only
> TEST 3.SM in my case)?

Yes

Sincerely,

Rémi
CHOQUET
 
Posts: 211
Joined: Thu Nov 24, 2005 4:58 am
Location: CEFE, Montpellier, FRANCE.


Return to analysis help

Who is online

Users browsing this forum: No registered users and 10 guests

cron