mpower0.1.0 package

Power Analysis via Monte Carlo Simulation for Correlated Data

bkmr_wrapper

Fits a BKMR model with significance criteria: PIP and group-wise PIP

bma_wrapper

Fits a linear model with Bayesian model selection with significance cr...

bws_wrapper

Fits a Bayesian weighted sums

cor2partial

Convert a correlation matrix into a partial correlation matrix

cvine

Citation: Daniel Lewandowski, Dorota Kurowicka, Harry Joe, Generating ...

estimate_snr

Monte Carlo approximation of the SNR

fin_wrapper

Fits a Bayesian factor model with interactions

fit

Fits the model to given data and gets a list of significance criteria

genx

Generates a matrix of n observations of p predictors

geny

Generates a vector of outcomes

glm_wrapper

Fits a generalized linear model

InferenceModel

Statistical model that returns significance criterion

MixtureModel

Correlated predictors generator

mplot

Visualize marginals and Gaussian copula correlations of simulated data

mpower

mpower: Power analysis using Monte Carlo for correlated predictors.

OutcomeModel

Outcome generator

partial

Partial correlations between elements in x and elements in y

plot_summary

Plot summaries of power simulation

qgcomp_lin_wrapper

Fits a linear Quantile G-Computation model with no interactions

qmultinom

Quantile function for the multinomial distribution, size = 1

rsq2snr

Convert R-squared value to the SNR

scale_f

Rescale the mean function of an OutcomeModel to meet a given SNR

scale_sigma

Rescale the noise variance of a Gaussian OutcomeModel to meet a given ...

set_value

This function updates values in an OutcomeModel object

sim_curve

Power curve using Monte Carlo simulation

sim_power

Power analysis using Monte Carlo simulation

summary

Tabular summaries of power simulation

A flexible framework for power analysis using Monte Carlo simulation for settings in which considerations of the correlations between predictors are important. Users can set up a data generative model that preserves dependence structures among predictors given existing data (continuous, binary, or ordinal). Users can also generate power curves to assess the trade-offs between sample size, effect size, and power of a design. This package includes several statistical models common in environmental mixtures studies. For more details and tutorials, see Nguyen et al. (2022) <arXiv:2209.08036>.

  • Maintainer: Phuc H. Nguyen
  • License: LGPL
  • Last published: 2022-09-21