fpc: Fishing Power Correction Factor from Experimental Fishing

In fishmethods: Fishery Science Methods and Models

Fishing Power Correction Factor from Experimental Fishing

Description

Calculates fishing power correction ratios between two vessels or gears

Usage

fpc(cpue1 = NULL, cpue2 = NULL, method = c(1,2,3,4),  deletezerosets = FALSE, 
kapp_zeros = "paired", boot_type = "paired", nboot = 1000, dint = c(1e-9,5),
 rint = c(1e-9, 20), decimals = 2, alpha = 0.05)

Arguments

cpue1	vector of CPUEs from vessel or gear considered the standard or baseline
cpue2	vector of CPUEs from other vessel or gear
method	method(s) to use to estimate fishing power correction. 1 = Ratio of Means, 2 = Randomized Block ANOVA, 3 = Multiplicative Model, 4 = Kappenman 1992. Default = c(1,2,3,4)
deletezerosets	if TRUE, paired observations with any CPUE=0 are eliminated prior to estimation. Default = FALSE.
kapp_zeros	for method = 4, how CPUE=0 is eliminated. "paired" eliminates the row of paired CPUE observations if CPUE = 0 is present for any observation within the pair, "ind" eliminates CPUE = 0 from the individual CPUE vectors.
boot_type	the method for bootstrapping data. "paired" = resample paired CPUE observations, "unpaired" = resample individual CPUE vectors
nboot	the number of bootstrap replicates. Default = 1000.
dint	the lower and upper limits of the function interval searched by function uniroot to solve Kappenman's d.
rint	the lower and upper limits of the function interval searched by function optimize to solve Kappenman's r.
decimals	the number of decimal places for output of estimates.
alpha	the alpha level used to calculate confidence intervals.

Details

The four methods for estimating fishing power correction factors given in Wilderbuer et al. (1998) are encoded.

If paired CPUE observations are both zero, the row is automatically eliminated. If deletezerosets = TRUE, the paired CPUE observations with any CPUE = 0 will be eliminated.

Zeroes are allowed in methods 1, 2 and 3.

For the Kappenman method (method=4), only non-zero CPUEs are allowed. Use kapp_zeros to select the elimination method. An unequal number of observations between vessels is allowed in this method and can result using kapp_zeros = "ind". FPC is derived by using the methodology where r that minimizes the sum of squares under the first conjecture relative to the second is estimated (Kappenman 1992: 2989; von Szalay and Brown 2001).

Standard errors and confidence intervals of FPC estimates are derived for most methods by using an approximation formula (where applicable), jackknifing and/or bootstrapping. Specify the type of bootstrapping through boot_type. For methods 1-3, jackknife estimates are provided only when boot_type="paired". If method = 4, jackknife estimates are provided only when boot_type="paired" and kapp_zeros="paired".

Confidence intervals are provided for the approximation formulae specified in Wilderbuer et al (1998), the jackknife estimates and bootstrap estimates. Confidence intervals for the jackknife method are calculated using the standard formula (estimate+/-z[alpha/2]*jackknife standard error). Bootstrap confidence intervals are derived using the percentile method (Haddon 2001).

Value

A dataframe containing method name, sample size for cpue1 (n1) and cpue2 (n2) ,mean cpue1, mean cpue2, fishing power correction (FPC), standard error from approximation formulae (U_SE), standard error from jackknifing (Jack_SE), standard error from bootstrapping (Boot_SE), lower and upper confidence intervals from approximation formulae (U_X%_LCI and U_X%_UCI),lower and upper confidence intervals from jackknifing (Jack_X%_LCI and Jack_X%_UCI) and lower and upper confidence intervals from bootstrapping (Boot_X%_LCI and Boot_X%_UCI).

Author(s)

Gary A. Nelson, Massachusetts Division of Marine Fisheries [e-mail]

References

Haddon, M. 2001. Modelling and Quantitative Methods in Fisheries. Chapman & Hall/CRC Press. Boca Raton, Florida.

Kappenman, R. F. 1992. Robust estimation of the ratio of scale parameters for positive random variables. Communications in Statistics, Theory and Methods 21: 2983-2996.

von Szalay, P. G. and E. Brown. 2001. Trawl comparisons of fishing power differences and their applicability to National Marine Fisheries Service and Alask Department of Fish and Game trawl survey gear. Alaska Fishery Research Bulletin 8(2):85-95.

Wilderbuer, T. K., R. F. Kappenman and D. R. Gunderson. 1998. Analysis of fishing power correction factor estimates from a trawl comparison experiment. North American Journal of Fisheries Management 18:11-18.

Examples

## Not run: 
 #FPC for flathead sole from von Szalay and Brown 2001
   data(sole)
   fpc(cpue1=sole$nmfs,cpue2=sole$adfg,boot_type="unpaired",kapp_zeros="ind",method=c(4),
            alpha=0.05)
## End(Not run)

fishmethods documentation built on April 27, 2023, 9:10 a.m.