Description Usage Arguments Value Author(s) References Examples
First fits selection curves by log-linear fitting using
glm
and then (if tangle is TRUE) uses the
SELECT method (of Millar 1992) to fit a slightly modified
model with a tangle parameter (see Bromaghin 2005 for the
concept of tangle parameters).
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 | fit.catch(catch, sel.curve = norm.loc, tangle = TRUE,
perimeter.factor = 1, tol = 1e-08, omega0 = 0.1, effort = NULL)
## S3 method for class 'fit.catch'
AIC(object, corrected = TRUE, ..., k = 2)
## S3 method for class 'fit.catch'
coef(object, ...)
## S3 method for class 'fit.catch'
coefficients(object, ...)
## S3 method for class 'fit.catch'
print(x, digits = max(3, getOption("digits") - 3), ...)
## S3 method for class 'fit.catch'
plot(x, y, xlab = "Fork length (mm)",
ylab = "Relative selection probability", resolution = 500,
plot.type = c("selcurve", "totalselcurve", "residuals",
"observedvrsexpected"), max.cex = 2, min.cex = 0.5, leg.pos = "topleft",
data.name = NULL, justtangle = TRUE, ...)
## S3 method for class 'fit.catch'
deviance(object, ...)
|
catch |
an object of class |
sel.curve |
one of a number of selection curve function including norm.loc, norm, lognorm, gamm and inv.gau. These functions create a range of functions that are required to make various calculations. In effect, the selection curve chosen defines the model to be fitted. |
tangle |
If TRUE, a tangle parameter is included in the fitted process (Bromaghin (2005). |
perimeter.factor |
Factor by which to multiply the inputted mesh sizes to obtain mesh perimeters, which are required by the analyses. perimeter.factor = 4; if using stretch mesh perimeter.factor = 2, the default). |
x |
an object of class |
object |
an object of class |
y |
not used. only for consistency with S3 plot method. |
xlab |
x-axis label |
ylab |
y-axis label |
resolution |
higher numbers make smoother selection curves |
plot.type |
type of plot to produce, either "selcurve", "totalselcurve", or "residuals" for selection curves, total selection curve, or residual plot; in this last type, the sizes of the circles are proportional to deviance |
max.cex |
maximum size of the deviance circles |
min.cex |
minimum size of the deviance circles |
leg.pos |
position of the legend if necessary, see
|
data.name |
name of the data set to be plotted |
justtangle |
should only the tangle model be plotted? |
corrected |
if TRUE, the small-sample-size corrected version of AIC is returned (recommended) |
k |
not used. only for consistency with default S3 method. |
tol |
tolerance used to decide on whether fitted values are numerically zero. used to correct AIC values. this correction depends on sample size, and the standard way to calculate sample size is as the number of mesh sizes times the number of fish length categories and then to subtract from this number the number of zero fitted counts |
omega0 |
initial value for the tangle parameter |
digits |
number of digits to round to |
... |
additional arguments to be passed |
effort |
TODO (sorry!) |
An object of class fit.catch
with components:
catch the original catch
object
sel.curve the original selection curve
theta the fitted parameters of the selection curve
mu the fitted values
res the deviance residuals (see Millar and Fryer 1999)
dev model deviance(s)
tangle TRUE if tangle parameter fitted, FALSE otherwise
call the matched call
l the 'length matrix'.
m the 'mesh matrix' (with input mesh sizes multiplied by
perimeter.factor
).
Steve Walker
J.F. Bromaghin (2005) A versatile net selectivity model, with application to Pacific salmon and freshwater species of the Yukon River, Alaska. Fisheries Research 74: 157-168.
R.B. Millar & R.J. Fryer (1999) Estimating the size-selection curves of towed gears, traps, nets and hooks. Reviews in Fish Biology and Fisheries 9: 89-116.
1 2 3 4 5 | data(north.pike)
np <- make.catch(north.pike$MESH, north.pike$FLEN)
plot(np)
(np.fit <- fit.catch(np, gamm))
plot(np.fit)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.