GEMACO, shortcuts and additivity

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

GEMACO, shortcuts and additivity

Postby simone77 » Mon Sep 03, 2012 6:18 am

Hi,

I am working in E-SURGE and I have a covariable with three levels (in the headed format input file).
1= males captured first time as juvenile
2= females captured first time as adults
3= males captured first time as adults

There are no females captured first time as juvenile, this is not strange for the biology of the species.
I have 3 states (healthy, sick, died) and four events because there are captures of individuals that I don’t know if they are healty or sick.

I have two transition steps, survival and, conditioned on survival, transition (healthy->sick and vice versa).

Question 1.

I have some trouble to write GEMACO sentences for additivity when the covariate does not come directly from the input file. For instance, how can I write this sentence in GEMACO for, say, survival?
“juvenile males”+(“sex of adults” + ”state” * “time”)
I.e. I want to get an estimate for juvenile males, and an additive effect between sex of adults and the interaction of state and time.

My attempt was to first create these shortcuts:
juvmal=[a(1).g(1)]
sexad=[g(2)+a(2).g(1)&g(3)] (I wanted to create a variable with two levels and did not find another way, the coma does not work in these cases)

and after write (for survival):

juvmal+[sexad+f.t]

But it doesn’t seem to work as I would, it seems that for each state it has done three parallel (additive effect) lines varying with time, one for juveniles males, one for adult females, and one for adult males.

Question 2.
In general, how can I keep track of who is who in the output file when I have used shortcuts? For instance in the previous sentence for survival I believed that juvmal was set as group 3 in the result file (given it was the only one without time effect - but it wasn't!). I don’t understand how it is assigned in the result file who is who.
simone77
 
Posts: 197
Joined: Mon Aug 10, 2009 2:52 pm

Re: GEMACO, shortcuts and additivity

Postby CHOQUET » Mon Sep 03, 2012 11:21 am

Hello,

Q1) It worked as it should. For a fixed state, you have biult an additive model between gender and time.
Q2) In the output file, there is no track (mark) in the results of shortcuts. This is a lack of E-SURGE.
To see who is who, you have the original group and the age from the first capture.

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

Re: GEMACO, shortcuts and additivity

Postby simone77 » Mon Sep 03, 2012 12:05 pm

Hi Rémi,

Thanks for your answer but please see below.

Q1) It worked as it should. For a fixed state, you have biult an additive model between gender and time.

No, it doesn't. I didn't explain it well, sorry, I want to get one estimate for juvmales (independent of time and state), and parallel estimates for adults according to their sex (for a fixed state).

Q2) In the output file, there is no track (mark) in the results of shortcuts. This is a lack of E-SURGE.
To see who is who, you have the original group and the age from the first capture.

OK. thanks
simone77
 
Posts: 197
Joined: Mon Aug 10, 2009 2:52 pm

Re: GEMACO, shortcuts and additivity

Postby CHOQUET » Tue Sep 04, 2012 3:41 am

For what you want, use the folllowing sentence:

juvmal+sexad(1:2).[sexad+f.t]


so that f.t applied only to sexad.
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 22 guests