[FLR-list] size

Wed May 22 17:17:42 CEST 2013

I think the discussion shows that we will need flexibility.

To try and summarise

For data objects
1) FLQuant can be used to model LFD, but not age as well
2) FLPar can be used to model say an ALK where you need both length & age

However we need to define what is meant by length in both cases, this could
be done by creating a new class (e.g. FLCohort) or adding an attribute
(e.g. M).

I think there are 3 types of length dims e,g,
I) a continuous variable, such as a mid pt which may come from a parametric
model
II) a variable that represents a bin boundary, either the left or righthand
boundary where the adjacent quant/param gives the other boundary
III) a variable that represents a bin boundary (l or r) and the bin width
is pre-defined, i.e. so you can have missing bins

Then methods and packages can be designed to work across packages.

Laurie

On Wed, May 22, 2013 at 4:18 PM, Laurie Kell <lauriekell at googlemail.com>wrote:

> It could be due to sparse data and some size classes being missing. For
> example going from a DB to a matrix.
>
> Laurie
>
>
> On Wed, May 22, 2013 at 4:05 PM, Ernesto <ernesto.jardim at jrc.ec.europa.eu>wrote:
>
>>  Hi,
>>
>> That can be done already, the range name we use for age is "max" and
>> "min" so we can use it directly for lengths.
>>
>> Not so sure about using different bin's size. Methods would need to check
>> each bin size ...
>>
>> One doubt, when using different bin's size aren't you getting close to
>> the age structure ? kind of slicing ?
>>
>> Best
>>
>> EJ
>>
>>
>> On 05/22/2013 03:38 PM, Santiago Cerviño wrote:
>>
>>  Length based models usually group length classes for different causes.
>> Sometimes you do not have the whole information; sometimes you need to
>> simplify your model; you might need to make big groups for larger fish and
>> small ones for medium size fish, etc. I think len bins should be user
>> defined without constraints. It is also important to have the whole
>> information (lower length and upper length) for each bin somewhere. This
>> information will be needed. ****
>>
>> ** **
>>
>> Could work this idea that keeps consistency with FLR age names using only
>> one number? For instance 5, 12, 20, 30 that means 5-12, 12-20, 20-30,
>> 30-??. We would only need that maximum length is defined somewhere to know
>> the size of the last bin (minage, maxage slots used for minlen maxlen).**
>> **
>>
>> ** **
>>
>> Santi****
>>
>> ** **
>>
>> ** **
>>  ------------------------------
>>
>> *De:* flr-list-bounces at lists.r-forge.r-project.org [
>> mailto:flr-list-bounces at lists.r-forge.r-project.org<flr-list-bounces at lists.r-forge.r-project.org>]
>> *En nombre de *Ernesto
>> *Enviado el:* 22 May 2013 13:55
>> *Para:* flr-list at r-forge.wu-wien.ac.at
>> *Asunto:* Re: [FLR-list] size****
>>
>> ** **
>>
>> It won't work in several situations and IMO trying to find a solution for
>> all doesn't work. We need to find a solution that works and implement
>> methods for the others. We may try to be consistent with "age" and use
>> [10,11) represented by "10" in the length names.
>>
>> Best
>>
>> EJ
>>
>> On 05/22/2013 02:43 PM, Laurie Kell wrote:****
>>
>>  but wont work with age slicing which is based on > & <=. ****
>>
>> Laurie****
>>
>> ** **
>>
>> On Wed, May 22, 2013 at 2:41 PM, Ernesto <ernesto.jardim at jrc.ec.europa.eu>
>> wrote:****
>>
>> Hi,
>>
>> Yes, that is a problem that we need to sort out. I'm considering using
>> midpoints. It has the advantage of being easy to coerce into numeric. In
>> most cases that's what one needs for computations. Using boundaries will
>> require other(s) operation(s) to get midpoints.
>>
>> Best
>>
>> EJ ****
>>
>>
>>
>> On 05/22/2013 02:23 PM, Laurie Kell wrote:****
>>
>>   There is an issue, boring but may be important to agree on now, i.e.
>> labeling of length bins.****
>>
>> What does a length bin mean, i.e. if the "quant" dim represents length?
>> and the values are 5, 10, ...?****
>>
>> In the mixdist package these are the righthand boundaries i.e. they
>> represent (0,5] and (5,10]. But if they are the lefthand boundaries then
>> they represent (5,10] and (10,INF]. Or they could be midpoints. One option
>> would be to explicitly label the bins with the boundaries and then have
>> methods for going backwards and forwards between parametric and
>> non-parametric methods. Or to have an attribute that tracks the properties.
>> ****
>>
>> Otherwise I can see a lot of time being wasted in stock assessments. well
>> more than normal.****
>>
>> Laurie ****
>>
>> ** **
>>
>> On Wed, May 22, 2013 at 1:07 PM, Ernesto <ernesto.jardim at jrc.ec.europa.eu>
>> wrote:****
>>
>> On 05/22/2013 12:02 PM, Ernesto wrote:****
>>
>> On 05/22/2013 11:34 AM, Sarah Kraak wrote:****
>>
>> Hi FLR-guys,
>>
>> As you know I am not an active FLR user, but last year I showed some
>> interest in whether you have size-based stuff in FLR (see attached email).
>> In the meanwhile Coilin Minto and myself have been working on something
>> (also attached - note this is a draft of the idea; at the moment we are
>> working on the code to get the final graphs out) which we are coding in R
>> but which could/should eventually been done in FLR. Perhaps you are
>> interested.
>>
>> Cheers,
>> Sarah****
>>
>>
>> Hi,
>>
>> Following Laurie's suggestion I've started a git repository for length
>> methods in FLRhttps://github.com/ejardim/FLMethLen
>>
>> Maybe our first task should be o write down what we're doing with length
>> data. Laurie's working on ALKs, so I guess you're interested in
>> converting length frequencies to ages for stock assessment !? We (a4a)
>> are working on OMs and how to generate them from length based data so
>> that uncertainty on the growth model can be included in the MSE. Sarah
>> and Colin Minto are working on selectivity. What else ?
>>
>> Best
>>
>> EJ****
>>
>> Hi,
>>
>> I've updated the readme to reflect the latest messages. ****
>>
>>
>>
>> EJ
>>
>> --
>> Ernesto Jardim<ernesto.jardim at jrc.ec.europa.eu>
>> Fisheries Scientist
>> FISHREG – Scientific Support to Fisheries
>> IPSC Maritime Affairs Unit
>> EC Joint Research Center
>> TP 051, Via Enrico Fermi 2749
>> I-21027 Ispra (VA), Italy
>> Office : +39 0332 785311 <%2B39%200332%20785311>
>> Fax: +39 0332 789658 <%2B39%200332%20789658>
>> https://fishreg.jrc.ec.europa.eu/
>>
>>
>> _______________________________________________
>> flr-list mailing list
>> flr-list at flr-project.org
>> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/flr-list****
>>
>> ** **
>>
>>
>>
>> ****
>>
>> _______________________________________________****
>>
>> flr-list mailing list****
>>
>> flr-list at flr-project.org****
>>
>> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/flr-list****
>>
>>
>>
>>
>> ****
>>
>> -- ****
>>
>> Ernesto Jardim<ernesto.jardim at jrc.ec.europa.eu> <ernesto.jardim at jrc.ec.europa.eu>****
>>
>> Fisheries Scientist****
>>
>> FISHREG – Scientific Support to Fisheries****
>>
>> IPSC Maritime Affairs Unit****
>>
>> EC Joint Research Center****
>>
>> TP 051, Via Enrico Fermi 2749****
>>
>> I-21027 Ispra (VA), Italy****
>>
>> Office : +39 0332 785311****
>>
>> Fax: +39 0332 789658****
>>
>> https://fishreg.jrc.ec.europa.eu/****
>>
>> ** **
>>
>>
>> _______________________________________________
>> flr-list mailing list
>> flr-list at flr-project.org
>> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/flr-list****
>>
>> ** **
>>
>>
>>
>>
>>  ****
>>
>> _______________________________________________****
>>
>> flr-list mailing list****
>>
>> flr-list at flr-project.org****
>>
>> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/flr-list****
>>
>>
>>
>>
>> ****
>>
>> -- ****
>>
>> Ernesto Jardim<ernesto.jardim at jrc.ec.europa.eu> <ernesto.jardim at jrc.ec.europa.eu>****
>>
>> Fisheries Scientist****
>>
>> FISHREG – Scientific Support to Fisheries****
>>
>> IPSC Maritime Affairs Unit****
>>
>> EC Joint Research Center****
>>
>> TP 051, Via Enrico Fermi 2749****
>>
>> I-21027 Ispra (VA), Italy****
>>
>> Office : +39 0332 785311****
>>
>> Fax: +39 0332 789658****
>>
>> https://fishreg.jrc.ec.europa.eu/****
>>
>> ** **
>>
>>
>>
>> _______________________________________________
>> flr-list mailing listflr-list at flr-project.orghttps://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/flr-list
>>
>>
>>
>> --
>> Ernesto Jardim<ernesto.jardim at jrc.ec.europa.eu> <ernesto.jardim at jrc.ec.europa.eu>
>> Fisheries Scientist
>> FISHREG – Scientific Support to Fisheries
>> IPSC Maritime Affairs Unit
>> EC Joint Research Center
>> TP 051, Via Enrico Fermi 2749
>> I-21027 Ispra (VA), Italy
>> Office : +39 0332 785311
>> Fax: +39 0332 789658https://fishreg.jrc.ec.europa.eu/
>>
>>
>> _______________________________________________
>> flr-list mailing list
>> flr-list at flr-project.org
>> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/flr-list
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/flr-list/attachments/20130522/d6a06aa2/attachment-0001.html>