params2ecdf: Parameter Values -> Empirical CDFs

In hzambran/hydroPSO: Particle Swarm Optimisation, with Focus on Environmental Models

Parameter Values -> Empirical CDFs

Description

This function computes (weighted) empirical CDFs (ECDFs) for each calibrated parameter, by using the parameter values obtained during the optimisation with hydroPSO (with optional plot)

Usage

params2ecdf(params, ...)

## Default S3 method:
params2ecdf(params, param.names=NULL, gofs=NULL, MinMax=NULL, 
        beh.thr=NA, weights=NULL, byrow=FALSE, plot=TRUE, obs=NULL, main=NULL, 
        nrows="auto", ylab="Probability", col="blue", leg.cex=1.2, 
        leg.pos="topleft", cex.axis=1.2, cex.main=1.2, cex.lab=1.2, 
        verbose=TRUE, ..., do.png=FALSE, png.width=1500, png.height=900, 
        png.res=90, png.fname="Params_ECDFs.png")
              
## S3 method for class 'matrix'
params2ecdf(params, param.names=colnames(params), gofs=NULL, 
        MinMax=NULL, beh.thr=NA, weights, byrow=FALSE, plot=TRUE, obs=NULL, 
        main=NULL, nrows="auto", ylab="Probability", col="blue", leg.cex=1.2, 
        leg.pos="topleft", cex.axis=1.2, cex.main=1.2, cex.lab=1.2, 
        verbose=TRUE, ..., do.png=FALSE, png.width=1500, png.height=900, 
        png.res=90, png.fname="Params_ECDFs.png")
              
## S3 method for class 'data.frame'
params2ecdf(params, param.names=colnames(params), gofs=NULL, 
        MinMax=NULL, beh.thr=NA, weights, byrow=FALSE, plot=TRUE, obs=NULL, 
        main=NULL, nrows="auto", ylab="Probability", col="blue", leg.cex=1.2, 
        leg.pos="topleft", cex.axis=1.2, cex.main=1.2, cex.lab=1.2, 
        verbose=TRUE, ..., do.png=FALSE, png.width=1500, png.height=900, 
        png.res=90, png.fname="Params_ECDFs.png")

Arguments

params	matrix or data.frame with the parameter values, where each row represent a different parameter set and each column represent the value of a different model parameter
param.names	character vector, names to be used for each parameter in params (by default its column names)
gofs	OPTIONAL. numeric with the values of goodness-of-fit values for each parameter in params (in the same order!)
MinMax	OPTIONAL. character, indicates if the optimum value in params corresponds to the minimum or maximum of the the objective function. Only used to identify the optimum in the plot Valid values are in: c('min', 'max')
beh.thr	numeric, used for selecting only the behavioural parameter sets, i.e. those with a goodness-of-fit value (as given in gofs) greater/less than or equal to beh.thr, depending on the value of MinMax By default beh.thr=NA and all the parameter sets are considered for the subsequent anlysis
weights	numeric vector, values of the weights to be used for computing the empirical CDFs Omitting the weights argument or specifying NULL or a zero-length vector will result in the usual un-weighted estimates
byrow	logical, indicates whether the computations have to be made for each column or for each row of params When the parameter sets are stored in rows, i.e. values for different model's parameter are stored in columns, byrow must be FALSE When the parameter sets are stored in columns, i.e. values for different model's parameter are stored in rows, byrow must be TRUE
plot	logical, indicates whether a plot with the Empirical CDF for each model's parameter has to be produced or not
obs	OPTIONAL. Only used when plot=TRUE Numeric or zoo object with observed values (one for each params), which are used in the output plot
main	an overall title for the plot
nrows	OPTIONAL. Only used when plot=TRUE numeric, number of rows to be used in the plotting window. If nrows is set to auto, the number of rows is automatically computed depending on the number of columns of params
ylab	OPTIONAL. Only used when plot=TRUE a title for the y axis. See plot
col	OPTIONAL. Only used when plot=TRUE a specification for the default plotting colour. See par
leg.cex	OPTIONAL. Only used when plot=TRUE character expansion factor *relative* to current 'par("cex")'. Used for text, and provides the default for 'pt.cex' and 'title.cex'. Default value = 1.2
leg.pos	OPTIONAL. Only used when plot=TRUE keyword to be used to position the legend. See legend
cex.axis	OPTIONAL. Only used when plot=TRUE numeric, magnification to be used for axis annotation relative to the current setting of cex
cex.main	OPTIONAL. Only used when plot=TRUE numeric, magnification to be used for main titles relative to the current setting of cex
cex.lab	OPTIONAL. Only used when plot=TRUE numeric, magnification to be used for x and y labels relative to the current setting of cex
verbose	logical, if TRUE, progress messages are printed
...	further arguments passed to the plot function or from other methods
do.png	logical, indicates if all the figures have to be saved into PNG files instead of the screen device
png.width	OPTIONAL. Only used when do.png=TRUE numeric with the width of the device. See png
png.height	OPTIONAL. Only used when do.png=TRUE numeric with the height of the device. See png
png.res	OPTIONAL. Only used when do.png=TRUE numeric with the nominal resolution in ppi which will be recorded in the PNG file, if a positive integer of the device. See png
png.fname	OPTIONAL. Only used when do.png=TRUE character, with the filename used to store the PNG file

Value

a list with nparam elements, where is the number of parameters given in params, computed as the number of columns in params.

Each element of the list contains the (weighted) empirical CDFs (ECDFs) for each parameter in params.

Additionally, if plot=TRUE a maximum of nfigs different figures with the Empirical CDF for each model's parameter is produced, where the maximum number of figures to be produced is computed as nfigs=ceiling(nparam/21).

Author(s)

Mauricio Zambrano-Bigiarini, [e-mail]

See Also

wtd.Ecdf, quant2ecdf

Examples

## Not run: 
# Setting the user's home directory as working directory
setwd("~")

# matrix with 100 random uniform parameter sets (in rows) for 10 different model's 
# parameters (in columns)
params           <- matrix(rnorm(1000), ncol=10, nrow=100)
colnames(params) <- paste("Param", 1:10, sep="")

# empirical CDFs for each one of the 10 parameters in 'params', with equal weight 
# for each one of the 100 parameter sets
params2ecdf(params, weights=rep(1,100)) 

## End(Not run)

hzambran/hydroPSO documentation built on Dec. 1, 2024, 3:28 p.m.