ARCHIVED

I'm archiving this repo - functions have been repackaged and moved to https://github.com/BenWilliams-NOAA/swo

SWO pseudocode

The sims function is the primary driver, it calls the pop_est function and replicates it a user defined number of iterations, it then processes the results out of list form and can save the results as a .csv.

pop_est(lfreq = length freq file, cpue = cpue file, samples = NULL, yrs = 2017, strata = NULL)

• yrs are the years to be sampled default >= 2017
• samples is the desired sample size by length
• strata is a switch to change the comps to year/strata instead of year change to TRUE if desired

1. get complete cases of lengths by species, year, and stratum = lngs a. there is a strata switch to include or exclude
2. filter years from lfreq and cpue data
3. calculate length comp a. no sampling 1. comps are calculated by year, species, stratum, haul, sex, and length b. reduced sample sizes by sex 1. filter lfreq for only males and females 2. uncount the frequency 3. sample by year, species, stratum, haul at the selected sample size, save as .new_sexed 4. filter out previously sexed fish that are now unsexed, .new_unsexed 5. Join the original unsexed fish from lfreq to the .new_sexed a. note: this means the overall sample size is reduced, could be computed in another fashion 6. comps are calculated by year, species, stratum, haul, sex, and length
4. calculate comp for hauls w/o length samples, .unk
5. id hauls w/o lengths, name .no_length
6. calculate the population estimate by stratum, .pop
7. if there are any hauls w/o lengths joins them to the length comp and compute population size by haul, .temp
8. if strata = TRUE aggregate abundance by year abd strata, save the output as new and the individuals as removed in a list a. if strata = NULL aggregate abundance by year and output directly to global environment

sims(iters = 1, lfreq = length freq file, cpue = cpue file, strata = NULL, samples = NULL, yrs = 2017, save = NULL)

• iters is the number of iterations
• strata is a switch to change the comps to year/strata instead of year change to TRUE if desired
• samples is the desired sample size by length
• years is the min year to sample aka >=
• save will save results in the output folder, a single file is output if there are no samples, otherwise two files are output: a comps file and a removed file (sexed individuals who are move dto the unsexed population)

1. replicates the pop_est function the desired number of iterations

get_data(data = sim output, id, species = NULL, yrs = NULL)

• id is the name of the data
• species is a filter
• yrs is a filter

1. helper function pulls the simulations, splits the results out calculates confidence intervals

plot_comp(base_data = og data, sim_data = reduced sample data, species = NULL, yrs = NULL)

• species is a filer
• yrs is a filter

1. line plot with 95% ci comparing original population-based comps to reduced sample size comps

table_comp(base_data = og data, sim_data = reduced sample data, species = NULL, yrs = NULL)

• species is a filer
• yrs is a filter

1. creates a table of output values, also output the difference and percent difference by year, or by year & stratum

plot_comp2(base_data, sim_data, species = NULL, yrs = NULL)

• species is a filer
• yrs is a filter

1. density plot to plot the output of table_comp