,
Alexander J. Jensen [aut] ,
Ryan P. Kelly [aut] ,
Andrew O. Shelton [aut] ,
William H. Satterthwaite [aut]
,
Eric C. Anderson [aut] | Title: | Bayesian Zero-and-One Inflated Dirichlet Regression Modelling |
|---|---|
| Description: | Fits Dirichlet regression and zero-and-one inflated Dirichlet regression with Bayesian methods implemented in Stan. These models are sometimes referred to as trinomial mixture models; covariates and overdispersion can optionally be included. |
| Authors: | Eric J. Ward [aut, cre] ,
Alexander J. Jensen [aut] ,
Ryan P. Kelly [aut] ,
Andrew O. Shelton [aut] ,
William H. Satterthwaite [aut]
,
Eric C. Anderson [aut] |
| Maintainer: | Eric J. Ward <[email protected]> |
| License: | GPL (>=3) |
| Version: | 1.3.1 |
| Built: | 2024-11-26 16:31:47 UTC |
| Source: | https://github.com/noaa-nwfsc/zoid |
A DESCRIPTION OF THE PACKAGE
Stan Development Team (2020). RStan: the R interface to Stan. R package version 2.21.2. https://mc-stan.org
Random generation of datasets using the dirichlet broken stick method
broken_stick( n_obs = 1000, n_groups = 10, ess_fraction = 1, tot_n = 100, p = NULL )broken_stick( n_obs = 1000, n_groups = 10, ess_fraction = 1, tot_n = 100, p = NULL )
n_obs |
Number of observations (rows of data matrix to simulate). Defaults to 10 |
n_groups |
Number of categories for each observation (columns of data matrix). Defaults to 10 |
ess_fraction |
The effective sample size fraction, defaults to 1 |
tot_n |
The total sample size to simulate for each observation. This is approximate and the actual simulated sample size will be slightly smaller. Defaults to 100 |
p |
The stock proportions to simulate from, as a vector. Optional, and when not included, random draws from the dirichlet are used |
A 2-element list, whose 1st element X_obs is the simulated dataset, and whose
2nd element is the underlying vector of proportions p used to generate the data
y <- broken_stick(n_obs = 3, n_groups = 5, tot_n = 100) # add custom proportions y <- broken_stick( n_obs = 3, n_groups = 5, tot_n = 100, p = c(0.1, 0.2, 0.3, 0.2, 0.2) )y <- broken_stick(n_obs = 3, n_groups = 5, tot_n = 100) # add custom proportions y <- broken_stick( n_obs = 3, n_groups = 5, tot_n = 100, p = c(0.1, 0.2, 0.3, 0.2, 0.2) )
Data from Satterthwaite, W.H., Ciancio, J., Crandall, E., Palmer-Zwahlen, M.L., Grover, A.M., O’Farrell, M.R., Anson, E.C., Mohr, M.S. & Garza, J.C. (2015). Stock composition and ocean spatial distribution from California recreational chinook salmon fisheries using genetic stock identification. Fisheries Research, 170, 166–178. The data genetic data collected from port-based sampling of recreationally-landed Chinook salmon in California from 1998-2002.
chinookchinook
A data frame.
Data from Magnussen, E. 2011. Food and feeding habits of cod (Gadus morhua) on the Faroe Bank. – ICES Journal of Marine Science, 68: 1909–1917. The data here are Table 3 from the paper, with sample proportions (columns w) multiplied by total weight to yield total grams (g) for each sample-diet item combination. Dashes have been replaced with 0s.
coddietcoddiet
A data frame.
Extract point estimates of compositions from fitted model.
fit_dirichlet(data)fit_dirichlet(data)
data |
The data to fit the dirichlet distribution to |
Find appropriate standard deviations for prior
fit_prior(n_bins, n_draws = 10000, target = 1/n_bins, iterations = 5)fit_prior(n_bins, n_draws = 10000, target = 1/n_bins, iterations = 5)
n_bins |
Bins for the Dirichlet distribution |
n_draws |
Numbers of samples to use for doing calculation |
target |
The goal of the specified prior, e.g. 1 or 1/n_bins |
iterations |
to try, to ensure robust solution. Defaults to 5 |
A 3-element list consisting of sd (the approximate standard deviation
in transformed space that gives a similar prior to that specified), value (the
value of the root mean squared percent error function being minimized),
and convergence (0 if convergence occurred, error code from
optim() otherwise)
# fit model with 3 components / alpha = 1 set.seed(123) f <- fit_prior(n_bins = 3, n_draws = 1000, target = 1) # fit model with 20 components / alpha = 1/20 f <- fit_prior(n_bins = 20, n_draws = 1000, target = 1 / 20)# fit model with 3 components / alpha = 1 set.seed(123) f <- fit_prior(n_bins = 3, n_draws = 1000, target = 1) # fit model with 20 components / alpha = 1/20 f <- fit_prior(n_bins = 20, n_draws = 1000, target = 1 / 20)
Fit a trinomial mixture model that optionally includes covariates to estimate effects of factor or continuous variables on proportions.
fit_zoid( formula = NULL, design_matrix, data_matrix, chains = 3, iter = 2000, warmup = floor(iter/2), overdispersion = FALSE, overdispersion_sd = 5, posterior_predict = FALSE, moment_match = FALSE, prior_sd = NA, ... )fit_zoid( formula = NULL, design_matrix, data_matrix, chains = 3, iter = 2000, warmup = floor(iter/2), overdispersion = FALSE, overdispersion_sd = 5, posterior_predict = FALSE, moment_match = FALSE, prior_sd = NA, ... )
formula |
The model formula for the design matrix. Does not need to have a response specified. If =NULL, then the design matrix is ignored and all rows are treated as replicates |
design_matrix |
A data frame, dimensioned as number of observations, and covariates in columns |
data_matrix |
A matrix, with observations on rows and number of groups across columns |
chains |
Number of mcmc chains, defaults to 3 |
iter |
Number of mcmc iterations, defaults to 2000 |
warmup |
Number iterations for mcmc warmup, defaults to 1/2 of the iterations |
overdispersion |
Whether or not to include overdispersion parameter, defaults to FALSE |
overdispersion_sd |
Prior standard deviation on 1/overdispersion parameter, Defaults to inv-Cauchy(0,5) |
posterior_predict |
Whether or not to return draws from posterior predictive distribution (requires more memory) |
moment_match |
Whether to do moment matching via |
prior_sd |
Parameter to be passed in to use as standard deviation of the normal distribution in transformed space. If covariates are included this defaults to 1, but for models with single replicate, defaults to 1/n_bins. |
... |
Any other arguments to pass to |
y <- matrix(c(3.77, 6.63, 2.60, 0.9, 1.44, 0.66, 2.10, 3.57, 1.33), nrow = 3, byrow = TRUE ) # fit a model with no covariates fit <- fit_zoid(data_matrix = y, chains = 1, iter = 100) # fit a model with 1 factor design <- data.frame("fac" = c("spring", "spring", "fall")) fit <- fit_zoid(formula = ~fac, design_matrix = design, data_matrix = y, chains = 1, iter = 100) # try a model with random effects set.seed(123) y <- matrix(runif(99,1,4), ncol=3) design <- data.frame("fac" = sample(letters[1:5], size=nrow(y), replace=TRUE)) design$fac <- as.factor(design$fac) fit <- fit_zoid(formula = ~(1|fac), design_matrix = design, data_matrix = y, chains = 1, iter = 100)y <- matrix(c(3.77, 6.63, 2.60, 0.9, 1.44, 0.66, 2.10, 3.57, 1.33), nrow = 3, byrow = TRUE ) # fit a model with no covariates fit <- fit_zoid(data_matrix = y, chains = 1, iter = 100) # fit a model with 1 factor design <- data.frame("fac" = c("spring", "spring", "fall")) fit <- fit_zoid(formula = ~fac, design_matrix = design, data_matrix = y, chains = 1, iter = 100) # try a model with random effects set.seed(123) y <- matrix(runif(99,1,4), ncol=3) design <- data.frame("fac" = sample(letters[1:5], size=nrow(y), replace=TRUE)) design$fac <- as.factor(design$fac) fit <- fit_zoid(formula = ~(1|fac), design_matrix = design, data_matrix = y, chains = 1, iter = 100)
Extract point estimates of compositions from fitted model.
get_fitted(fitted_model, conf_int = 0.05)get_fitted(fitted_model, conf_int = 0.05)
fitted_model |
The fitted model returned as an rstan object from the call to zoid |
conf_int |
Parameter controlling confidence intervals calculated, defaults to 0.05 for 95% intervals |
A list containing the posterior summaries of estimated parameters, with
element mu (the predicted values in normal space). For predictions
in transformed space, or overdispersion, see get_pars()
y <- matrix(c(3.77, 6.63, 2.60, 0.9, 1.44, 0.66, 2.10, 3.57, 1.33), nrow = 3, byrow = TRUE ) # fit a model with no covariates fit <- fit_zoid(data_matrix = y) p_hat <- get_fitted(fit)y <- matrix(c(3.77, 6.63, 2.60, 0.9, 1.44, 0.66, 2.10, 3.57, 1.33), nrow = 3, byrow = TRUE ) # fit a model with no covariates fit <- fit_zoid(data_matrix = y) p_hat <- get_fitted(fit)
Extract estimated parameters from fitted model.
get_pars(fitted_model, conf_int = 0.05)get_pars(fitted_model, conf_int = 0.05)
fitted_model |
The fitted model returned as an rstan object from the call to zoid |
conf_int |
Parameter controlling confidence intervals calculated, defaults to 0.05 for 95% intervals |
A list containing the posterior summaries of estimated parameters. At minimum,
this will include p (the estimated proportions) and betas (the predicted values in
transformed space). For models with overdispersion, an extra
element phi will also be returned, summarizing overdispersion. For models with random
intercepts, estimates of the group level effects will also be returned as zetas (again,
in transformed space). For predictions
in normal space, see get_fitted()
y <- matrix(c(3.77, 6.63, 2.60, 0.9, 1.44, 0.66, 2.10, 3.57, 1.33), nrow = 3, byrow = TRUE ) # fit a model with no covariates fit <- fit_zoid(data_matrix = y) p_hat <- get_pars(fit)y <- matrix(c(3.77, 6.63, 2.60, 0.9, 1.44, 0.66, 2.10, 3.57, 1.33), nrow = 3, byrow = TRUE ) # fit a model with no covariates fit <- fit_zoid(data_matrix = y) p_hat <- get_pars(fit)
Fit a trinomial mixture model that optionally includes covariates to estimate effects of factor or continuous variables on proportions.
parse_re_formula(formula, data)parse_re_formula(formula, data)
formula |
The model formula for the design matrix. |
data |
The data matrix used to construct RE design matrix |
Extract point estimates of compositions from fitted model.
rmspe_calc(par, n_bins, n_draws, target)rmspe_calc(par, n_bins, n_draws, target)
par |
The parameter (standard deviation) to be searched over to find a Dirichlet equivalent |
n_bins |
Bins for the Dirichlet distribution |
n_draws |
Numbers of samples to use for doing calculation |
target |
The goal of the specified prior, e.g. 1 or 1/n_bins |