blocks: Estimating a block model for sequential data

In roliveros-ramos/sharkbox: Shark assessment toolbox

Description

block is used to estimate a 'block model' (REFERENCE) based on the dominant level of a variable in a given ordered sequence.

Usage

blocks(formula, data, thr = 0.03, verbose = TRUE, na.action,
  method = "default", ...)

Arguments

formula	an object of class "formula" (or one that can be coerced to that class): a symbolic description of the model to be fitted. The details of model specification are given under 'Details'.
data	an optional data frame, list or environment (or object coercible by as.data.frame to a data frame) containing the variables in the model. If not found in data, the variables are taken from environment(formula), typically the environment from which glm is called.
thr	Threshold for the length of a block in order to be identified.
verbose	a logical indicating if some "progress report" should be given.
na.action	a function which indicates what should happen when the data contain NAs. The default is set by to na.pass.
method	The method used for reduction of groups. See 'Details'.
...	Not currently used, just in case.

Details

We define an "unloading block" as a group of fish that were predominantly of the same type (e.g. species, quality, size class) within each individual unloading. Unloading blocks are allowed to contain some minor amount of fish with different characteristics for practical reasons. To estimate the unloading blocks in each super-sample, the n-running proportion for each type of fish in the unloading is calculated and the dominant type (more than 50 fishes. An unloading block is then computed as the union of contiguous groups with the same dominant type. Small blocks (less than n fish unloaded) dividing two blocks of the same dominant type are absorved to generate uninterrupted unloading blocks of the same type. Since the expected length of a unloading block is expected to change as function of the total unloading length (i.e. the longer the unloading the longer the unloading blocks), the value n to compute the running proportions should be taken as a fraction (e.g. 3%, thr=0.03) of the total number fish unloaded N.

Examples

## Not run: 

data_file = system.file("data/supersamples_demo.csv", package="sharkbox")
dat = read.csv(data_file)
mod_block0 = blocks(order|trip ~ species, data=dat)
mod_block1 = blocks(order|trip ~ group, data=dat)

## End(Not run)

roliveros-ramos/sharkbox documentation built on Oct. 4, 2019, 1:36 p.m.