Learn R Programming

propagate: Propagation of uncertainty using higher-order Taylor expansion and Monte Carlo simulation (R-package)

propagate is an R package that can conduct uncertainty propagation based on first- and second-order Taylor expansion as well as Monte Carlo simulation. It also houses functionality to estimate confidence and prediction intervals for nonlinear models, create large correlation matrices and automatic distribution fitting.

Installation

You can install the latest development version of the code using the devtools R package.

# Install devtools, if you haven't already.
install.packages("devtools")
library(devtools)
install_github("anspiess/propagate")
source("https://install-github.me/anspiess/propagate")

Copy Link

Version

Install

install.packages('propagate')

Monthly Downloads

758

Version

1.2-0

License

GPL (>= 2)

Maintainer

Andrej-Nikolai Spiess

Last Published

July 27th, 2026

Functions in propagate (1.2-0)

statVec

Transform an input vector into one with defined mean and standard deviation
sobol

Compute Sobol Sensitivity Indices from a propagate object
propagate

Propagation of uncertainty using higher-order Taylor expansion and Monte Carlo simulation
summary.propagate

Summary function for 'propagate' objects
moments

Skewness and (excess) Kurtosis of a vector of values
numDerivs

Functions for creating Gradient and Hessian matrices by numerical differentiation (Richardson's method) of the partial derivatives
rDistr

Creating random samples from a variety of useful distributions
plot.propagate

Plotting function for 'propagate' objects
stochContr

Stochastic contribution analysis of Monte Carlo simulation-derived propagated uncertainty
predictNLS

Confidence/prediction intervals for (weighted) nonlinear models based on uncertainty propagation
WelchSatter

Welch-Satterthwaite approximation to the 'effective degrees of freedom'
matrixStats

Fast column- and row-wise versions of variance coded in C++
bigcor

Creating very large correlation/covariance matrices
makeDat

Create a dataframe from the variables defined in an expression
makeDerivs

Utility functions for creating Gradient- and Hessian-like matrices with symbolic derivatives and evaluating them in an environment
cor2cov

Converting a correlation matrix into a covariance matrix
fitDistr

Fitting distributions to observations/Monte Carlo simulations
mixCov

Aggregating covariances matrices and/or error vectors into a single covariance matrix
interval

Uncertainty propagation based on interval arithmetics
datasets

Datasets from the GUM "Guide to the expression of uncertainties in measurement" (2008)