character weights

[Chuck Mitchell]

I am trying to manually set character weights for a couple characters
as part of a sensitivity analysis. I have tried several options using
syntax of the CCODE command in the data file as best I understand it,
but all have failed.

As I understand it, the basic setting is
;
ccode
/1-[0
/1-[2
...etc.

with /N (1 in this case) as the weight. So, I changed the weights for
the four possibly correlated characters to 0.25 so that they would
collectively have the same weight as a single character weighted 1
with the syntax:

/0.25-[0

This generates an error -- 0 weight is not allowed. I next tried:

/.25[0

but that did nothing. The analysis ran but the trees had the same
length as when all characters were weight 1.

I infer from this that TNT does not support fractional weights. Is
that correct, or have I got the syntax wrong?

Since I have another set of 5 characters that I want collectively to
have the same weight as other single characters, I tried whole number
weights of 4, 5, and 20, for the set of five, the set of four, and
the rest, respectively, but the aquickie.run script crashes with a
memory error and a message about increasing slack for sectors to 20 or
more... I don't know how to read or write scripts and cannot find
enough info in the TNT wiki and web site to figure out how to fix
this.

Does anyone have a simple suggestion about how to achieve this
sensitivity analysis?

[Santiago Catalano]

Hi Chuck,

1-. TNT does not accept fractional weights. What you have to do instead is to upweight the rest of the characters. 

2-. In order to avoid the memory error, before running the script type in the command line   

 sect  : slack 30 ;  (or a higher value) 

Santiago

 

--
Has recibido este mensaje porque estás suscrito al grupo "TNT-Tree Analysis using New Technology" de Grupos de Google.
Para publicar una entrada en este grupo, envía un correo electrónico a tnt-tree-analysis-u...@googlegroups.com.
Para anular tu suscripción a este grupo, envía un correo electrónico a tnt-tree-analysis-using-n...@googlegroups.com
Para tener acceso a más opciones, visita el grupo en http://groups.google.com/group/tnt-tree-analysis-using-new-technology?hl=es.

--
Dr. Santiago A. Catalano
Fundación Miguel Lillo - INSUE 

CONICET
SM Tucumán - Tucumán
Argentina

[Charles Mitchell]

That worked!  Thank you. 

-- Cheers, Chuck

[Martha Calderon]

Dear Santiago,

I was checking this point, I wonder if there is a maximum value for
upweight the characters or not. Maybe, is it depending on my
characters number?. Thanks for your time. I will appreciate your
answer.

Best regards

Martha

> >>http://groups.google.com/group/tnt-tree-analysis-using-new-technology...
> >> .

>
> --
> Dr. Santiago A. Catalano
> Fundación Miguel Lillo - INSUE
> CONICET
> SM Tucumán - Tucumán

> Argentina- Ocultar texto de la cita -
>
> - Mostrar texto de la cita -

[Santiago Catalano]

Hi Martha,

The highest weight you can give is 1000.

bests

S.

Para tener acceso a más opciones, visita el grupo en http://groups.google.com/group/tnt-tree-analysis-using-new-technology?hl=es.