Package 'PEcAn.priors' reference manual

Title:	PEcAn Functions Used to Estimate Priors from Data
Description:	Functions to estimate priors from data.
Authors:	David LeBauer
Maintainer:	David LeBauer <dlebauer@illinois.edu>
License:	BSD_3_clause + file LICENSE
Version:	1.7.3.9000
Built:	2024-08-15 14:16:51 UTC
Source:	https://github.com/PecanProject/pecan

Create Density Data Frame from Sample

Description

Returns a data frame from stats::density function

Usage

create.density.df(
  samps = NULL,
  zero.bounded = FALSE,
  distribution = NULL,
  n = 1000,
  ...
)
create.density.df(
  samps = NULL,
  zero.bounded = FALSE,
  distribution = NULL,
  n = 1000,
  ...
)

Arguments

`samps`	a vector of samples from a distribution
`zero.bounded`
`distribution`	list with elements distn, parama, paramb, e.g. `list('norm', 0, 1)`

Value

data frame with x and y = dens(x)

Author(s)

David LeBauer

Examples

prior.df <- create.density.df(distribution = list('norm',0,1))
plot(prior.df)
samp.df <- create.density.df(samps = rnorm(100))
lines(samp.df)
prior.df <- create.density.df(distribution = list('norm',0,1))
plot(prior.df)
samp.df <- create.density.df(samps = rnorm(100))
lines(samp.df)

Get the quantiles from prior density

Description

Find quantiles on a density data frame

Usage

get.quantiles.from.density(density.df, quantiles = c(0.025, 0.5, 0.975))
get.quantiles.from.density(density.df, quantiles = c(0.025, 0.5, 0.975))

Arguments

`quantiles`	default is the median and 95% CI;
`priordensity`	density dataframe generated by `create.density.df`

Author(s)

David LeBauer

Examples

prior.df <- create.density.df(distribution = list('norm',0,1))
get.quantiles.from.density(prior.df)
samp.df <- create.density.df(samps = rnorm(100))
get.quantiles.from.density(samp.df)
prior.df <- create.density.df(distribution = list('norm',0,1))
get.quantiles.from.density(prior.df)
samp.df <- create.density.df(samps = rnorm(100))
get.quantiles.from.density(samp.df)

Get Samples

Description

Take n random samples from prior

Usage

get.sample(prior, n = NULL, p = NULL)
get.sample(prior, n = NULL, p = NULL)

Arguments

`prior`	data.frame with distn, parama, and optionally paramb.
`n`	number of samples to return from a random sample of the rdistn family of functions (e.g. rnorm)
`p`	vector of quantiles from which to sample the distribution; typically pre-generated upstream in the workflow to be used by the qdistn family of functions (e.g. qnorm)

Details

Similar to the prior sample function pr.samp, except 1) it takes the prior as a named dataframe or list and it can return either a random sample of length n OR a sample from a quantile specified as p

Value

vector with n random samples from prior

Examples

## Not run: 
# return 1st through 99th quantile of standard normal distribution:
PEcAn.priors::get.sample(
   prior = data.frame(distn = 'norm', parama = 0, paramb = 1), 
   p = 1:99/100)
# return 100 random samples from standard normal distribution:
PEcAn.priors::get.sample(
   prior = data.frame(distn = 'norm', parama = 0, paramb = 1), 
   n = 100)

## End(Not run)
## Not run: 
# return 1st through 99th quantile of standard normal distribution:
PEcAn.priors::get.sample(
   prior = data.frame(distn = 'norm', parama = 0, paramb = 1), 
   p = 1:99/100)
# return 100 random samples from standard normal distribution:
PEcAn.priors::get.sample(
   prior = data.frame(distn = 'norm', parama = 0, paramb = 1), 
   n = 100)

## End(Not run)

Plot probability density and data

Description

Plot probability density and data

Usage

plot_densities(density.plot_inputs, outdir, ...)
plot_densities(density.plot_inputs, outdir, ...)

Arguments

`density.plot_inputs`	list containing trait.samples and trait.df
`outdir`	directory in which to generate figure as pdf
`...`	passed on to plot_density

Value

outputs plots in outdir/sensitivity.analysis.pdf file

Author(s)

David LeBauer

Add posterior density to a plot

Description

Add posterior density to a plot

Usage

plot_posterior.density(posterior.density, base.plot = NULL)
plot_posterior.density(posterior.density, base.plot = NULL)

Arguments

`posterior.density`	data frame containing columns x and y
`base.plot`	a ggplot object (grob), created if none provided

Value

plot with posterior density line added

Author(s)

David LeBauer

Plots a prior density from a parameterized probability distribution

Description

Plots a prior density from a parameterized probability distribution

Usage

plot_prior.density(prior.density, base.plot = NULL, prior.color = "black")
plot_prior.density(prior.density, base.plot = NULL, prior.color = "black")

Arguments

`prior.density`	data frame containing columns x and y
`base.plot`	a ggplot object (grob), created if none provided
`prior.color`	color of line to be plotted

Value

plot with prior density added

Author(s)

David LeBauer

Examples

## Not run: 
plot_prior.density(pr.dens('norm', 0, 1))

## End(Not run)
## Not run: 
plot_prior.density(pr.dens('norm', 0, 1))

## End(Not run)

Plot trait density and data

Description

Plot trait density and data

Usage

plot_trait(
  trait,
  prior = NULL,
  posterior.sample = NULL,
  trait.df = NULL,
  fontsize = list(title = 18, axis = 14),
  x.lim = NULL,
  y.lim = NULL,
  logx = FALSE
)
plot_trait(
  trait,
  prior = NULL,
  posterior.sample = NULL,
  trait.df = NULL,
  fontsize = list(title = 18, axis = 14),
  x.lim = NULL,
  y.lim = NULL,
  logx = FALSE
)

Arguments

`trait`	dataframe with id, figid and units of the trait to be plotted
`prior`	named distribution with parameters
`posterior.sample`	samples from posterior distribution whose density should be plotted
`trait.df`	data to be plotted, in a format accepted by `jagify`
`fontsize`, `x.lim`, `y.lim`, `logx`	passed on to ggplot

Value

plot (grob) object

Author(s)

David LeBauer

Examples

## Not run: 
prior1 <- data.frame(distn = 'norm',
                     parama = 20,
                     paramb = 5)
data1  <- data.frame(Y = c(19, 21), se = c(1,1))
trait1 <- data.frame(id = 'Vcmax', figid = 'Vcmax', units = 'umol CO2 m-2 s-1')
plot_trait(trait = trait1,
          prior = prior1,
          trait.df = data1)

## End(Not run)
## Not run: 
prior1 <- data.frame(distn = 'norm',
                     parama = 20,
                     paramb = 5)
data1  <- data.frame(Y = c(19, 21), se = c(1,1))
trait1 <- data.frame(id = 'Vcmax', figid = 'Vcmax', units = 'umol CO2 m-2 s-1')
plot_trait(trait = trait1,
          prior = prior1,
          trait.df = data1)

## End(Not run)

Calculate densities

Description

Calculates density at n points across the range of a parameter

Usage

pr.dens(distn, parama, paramb, n = 1000, alpha = 0.001)
pr.dens(distn, parama, paramb, n = 1000, alpha = 0.001)

Arguments

`distn`	distribution
`parama`	parameter
`paramb`	parameter
`n`	length of vector to be returned
`alpha`	sets range at which the distribution will be evaluated (e.g. from alpha to 1-alpha)

Details

For a distribution and parameters, return the density for values ranging from alpha to 1-alpha

Value

dataframe with equally spaced x values and the corresponding densities

Author(s)

David LeBauer

prior.fn

Description

Prior fitting function for optimization

Usage

prior.fn(parms, x, alpha, distn, central.tendency = NULL, trait = NULL)
prior.fn(parms, x, alpha, distn, central.tendency = NULL, trait = NULL)

Arguments

`parms`	target for optimization
`x`	vector with c(lcl, ucl, ct) lcl / ucl = confidence limits, ct = entral tendency
`alpha`	quantile at which lcl/ucl are estimated (e.g. for a 95% CI, alpha = 0.5)
`distn`	named distribution, one of 'lnorm', 'gamma', 'weibull', 'beta'; support for other distributions not currently implemented
`central.tendency`	one of 'mode', 'median', and 'mean'
`trait`	name of trait, can be used for exceptions (currently used for trait == 'q')

Details

This function is used within DEoptim to parameterize a distribution to the central tendency and confidence interval of a parameter. This function is not very robust; currently it needs to be tweaked when distributions require starting values (e.g. beta, f)

Value

parms

Author(s)

David LeBauer

Examples

## Not run: 
  DEoptim(fn = prior.fn,
          lower = c(0, 0),
          upper = c(1000, 1000),
          x=c(2, 6, 3.3),
          alpha = 0.05,
          distn = 'lnorm')$optim$bestmem

## End(Not run)

## Not run: 
  DEoptim(fn = prior.fn,
          lower = c(0, 0),
          upper = c(1000, 1000),
          x=c(2, 6, 3.3),
          alpha = 0.05,
          distn = 'lnorm')$optim$bestmem

## End(Not run)

Prior Figure

Description

Plot prior density and data

Usage

priorfig(
  priordata = NA,
  priordensity = NA,
  trait = NA,
  xlim = "auto",
  fontsize = 18
)
priorfig(
  priordata = NA,
  priordensity = NA,
  trait = NA,
  xlim = "auto",
  fontsize = 18
)

Arguments

`priordata`	observations to be plotted as points
`priordensity`	density of prior distribution, calculated by `prior.density`
`trait`	dataframe with id, figid and units of the trait
`xlim`	limits for x axis

Value

plot / grob of prior distribution with data used to inform the distribution

Author(s)

David LeBauer

`trait.data`	data for distribution
`dists`	list of distribution names

Package 'PEcAn.priors'

Help Index

Create Density Data Frame from Sample

Description

Usage

Arguments

Value

Author(s)

Examples

Fit distribution to data

Description

Usage

Arguments

Value

Author(s)

Get the quantiles from prior density

Description

Usage

Arguments

Author(s)

Examples

Get Samples

Description

Usage

Arguments

Details

Value

See Also

Examples

Plot probability density and data

Description

Usage

Arguments

Value

Author(s)

Add posterior density to a plot

Description

Usage

Arguments

Value

Author(s)

Plots a prior density from a parameterized probability distribution

Description

Usage

Arguments

Value

Author(s)

See Also

Examples

Plot trait density and data

Description

Usage

Arguments

Value

Author(s)

Examples

Calculate densities

Description

Usage

Arguments

Details

Value

Author(s)

Sample from prior

Description

Usage

Arguments

Value

See Also

prior.fn

Description

Usage

Arguments

Details

Value

Author(s)

Examples

Prior Figure

Description

Usage