Description Usage Arguments Value References Examples
Format for, fit, and predict from a panel regression model. The
datalimited package includes cached data that can be used with
fit_prm
and a cached model that can be used with predict_prm
.
These functions can also be used with new data.
format_prm
: Format a series of predictors for a panel
regression
fit_prm
: Fit a panel regression
predict_prm
: Predict from a panel regression model
1 2 3 4 5 6 7 8 | format_prm(year, catch, species_cat, bbmsy = NULL)
fit_prm(dat, eqn = log(bbmsy) ~ max_catch + mean_scaled_catch + scaled_catch +
scaled_catch1 + scaled_catch2 + scaled_catch3 + scaled_catch4 + species_cat +
catch_to_rolling_max + time_to_max + years_back + initial_slope - 1,
type = c("lm", "gbm"), ...)
predict_prm(newdata, model = datalimited::ram_prm_model, ci = FALSE)
|
year |
A numeric vector of years |
catch |
A numeric vector of catches |
species_cat |
A single character value of a species category. Can be from any set of species categories as long as the same set are used in model fitting and extrapolation. |
bbmsy |
A numeric vector of B/B_MSY (only needed if passing the output
to |
dat |
A data frame created by |
eqn |
A formula describing the regression |
type |
Either linear model ( |
... |
Anything extra to pass to |
newdata |
A data frame to predict on that has been formatted with
|
model |
A linear model to predict from built from |
ci |
Should confidence intervals on B/B_MSY be returned? |
format_prm
: A data frame formatted for use with fit_prm
or
predict_prm
fit_prm
: A linear model for use with predict_prm
predict_prm
: If ci = FALSE
, a vector of predictions of
B/B_MSY. If ci = TRUE
, a data frame.
Note that the model is fitted to log(B/B_MSY) and the output
from predict_prm
is exponentiated so the prediction represents an
estimate of median B/B_MSY.
Costello, C., D. Ovando, R. Hilborn, S. D. Gaines, O. Deschenes, and S. E. Lester. 2012. Status and Solutions for the World's Unassessed Fisheries. Science 338:517-520.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 | # combine two built in datasets:
# d <- dplyr::inner_join(ram_ts, spp_categories, by = "scientificname")
# ram_prm_dat is built into the package; it can be created with:
# ram_prm_dat <- plyr::ddply(d, "stockid", function(x) {
# format_prm(year = x$year, catch = x$catch, bbmsy = x$bbmsy_ram,
# species_cat = x$spp_category[1L])
# })
# ram_prm_model is built into the package, it is created in the same manner
# as this:
# m <- fit_prm(ram_prm_dat)
# now predict B/Bmsy:
d <- subset(ram_prm_dat, stockid == "BGRDRSE")
x <- predict_prm(d) # use the built in model
plot(x)
# with confidence intervals:
library("ggplot2")
x <- predict_prm(d, ci = TRUE)
ggplot(x, aes(year, bbmsy_q50)) +
geom_ribbon(aes(ymin = bbmsy_q2.5, ymax = bbmsy_q97.5), fill = "#00000040") +
geom_line() +
ylab(expression(B/B[MSY]))
# format and predict on a new fake dataset:
set.seed(1)
dat <- format_prm(
year = 1:10,
catch = rlnorm(10),
species_cat = "Cods, hakes, haddocks")
head(dat)
predict_prm(dat)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.