Package index • BGGM

Missing Data

Handle Missing Values.

bggm_missing(): GGM: Missing Data

impute_data(): Obtain Imputed Datasets

Estimation Based Methods

‘Estimation’ indicates that the methods to not employ Bayes factor testing. Rather, the graph is determined with the posterior distribution. The prior distribtuion has a minimal influence.

estimate(): GGM: Estimation

coef(<estimate>): Compute Regression Parameters for estimate Objects

predict(<estimate>): Model Predictions for estimate Objects

plot(<summary.estimate>): Plot summary.estimate Objects

select(<estimate>): Graph Selection for estimate Objects

summary(<estimate>): Summary method for estimate.default objects

Exploratory Hypothesis Testing

Bayes factor testing to determine the graph. ‘Exploratory’ reflects that there is not a specific hypothesis being test.

explore(): GGM: Exploratory Hypothesis Testing

coef(<explore>): Compute Regression Parameters for explore Objects

predict(<explore>): Model Predictions for explore Objects

plot(<summary.explore>): Plot summary.explore Objects

plot(<summary.select.explore>): Plot summary.select.explore Objects

select(<explore>): Graph selection for explore Objects

summary(<explore>): Summary Method for explore.default Objects

summary(<select.explore>): Summary Method for select.explore Objects

Confirmatory Hypothesis Testing

Test (in)equality constrained hypotheses with the Bayes factor.

confirm(): GGM: Confirmatory Hypothesis Testing

plot(<confirm>): Plot confirm objects

Compare Gaussian Graphical Models

A variety of methods for comparing GGMs.

Posterior Predictive Check

Compare groups with a posterior predictive check, where the null model is that the groups are equal. This works with any number of groups. There is also an option to compare the groups with a user defined test-statistic.

ggm_compare_ppc(): GGM Compare: Posterior Predictive Check

plot(<ggm_compare_ppc>): Plot ggm_compare_ppc Objects

Partial Correlation Differences

Pairwise comparisons for each partial correlation in the respective models. This can be used for any number of groups. There is also an analytical solution.

ggm_compare_estimate(): GGM Compare: Estimate

plot(<summary.ggm_compare_estimate>): Plot summary.ggm_compare_estimate Objects

select(<ggm_compare_estimate>): Graph Selection for ggm_compare_estimate Objects

summary(<ggm_compare_estimate>): Summary method for ggm_compare_estimate objects

Exploratory Hypothesis Testing

Pairwise comparisions with exploratory hypothesis testing. This method can be used to compare several groups simultaneously.

ggm_compare_explore(): GGM Compare: Exploratory Hypothesis Testing

plot(<summary.ggm_compare_explore>): Plot summary.ggm_compare_explore Objects

select(<ggm_compare_explore>): Graph selection for ggm_compare_explore Objects

summary(<ggm_compare_explore>): Summary Method for ggm_compare_explore Objects

Confirmatory Hypothesis Testing

Test (in)equality constrained hypotheses with the Bayes factor.

ggm_compare_confirm(): GGM Compare: Confirmatory Hypothesis Testing

plot(<confirm>): Plot confirm objects

Predictability

Bayesian variance explained for each node in the model.

predictability(): Predictability: Bayesian Variance Explained (R2)

plot(<predictability>): Plot predictability Objects

summary(<predictability>): Summary Method for predictability Objects

Network Statistics

Compute network statistics from a partial correlation matrix or a weighted adjacency matrix.

roll_your_own(): Compute Custom Network Statistics

plot(<roll_your_own>): Plot roll_your_own Objects

Partial Correlation Sums

Compute the sum of partial correlations within (one group) or between (two groups) GGMs. This can be used to compare sums.

pcor_sum(): Partial Correlation Sum

plot(<pcor_sum>): Plot pcor_sum Object

Network Plot

Network plot for the selected graphs. This works with all method for which there is a selected graph.

plot(<select>): Network Plot for select Objects

Graphical VAR (vector autoregression)

A variety of methods for time series data. These particular models are VAR(1) models which are also known as time series chain graphical models.

Estimation

‘Estimation’ indicates that the methods to not employ Bayes factor testing. Rather, the graph is determined with the posterior distribution. The prior distribtuion has a minimal influence.

var_estimate(): VAR: Estimation

select(<var_estimate>): Graph Selection for var.estimate Object

summary(<var_estimate>): Summary Method for var_estimate Objects

plot(<summary.var_estimate>): Plot summary.var_estimate Objects

predict(<var_estimate>): Model Predictions for var_estimate Objects

Miscellaneous

convergence(): MCMC Convergence

fisher_z_to_r(): Fisher Z Back Transformation

fisher_r_to_z(): Fisher Z Transformation

gen_ordinal(): Generate Ordinal and Binary data

pcor_to_cor(): Compute Correlations from the Partial Correlations

pcor_mat(): Extract the Partial Correlation Matrix

plot_prior(): Plot: Prior Distribution

posterior_samples(): Extract Posterior Samples

map(): Maximum A Posteriori Precision Matrix

regression_summary(): Summarary Method for Multivariate or Univarate Regression

summary(<coef>): Summarize coef Objects

weighted_adj_mat(): Extract the Weighted Adjacency Matrix

zero_order_cors(): Zero-Order Correlations

Data

Example datasets and correlation matrices.

asd_ocd: Data: Autism and Obssesive Compulsive Disorder

bfi: Data: 25 Personality items representing 5 factors

csws: Data: Contingencies of Self-Worth Scale (CSWS)

depression_anxiety_t1: Data: Depression and Anxiety (Time 1)

depression_anxiety_t2: Data: Depression and Anxiety (Time 2)

gss: Data: 1994 General Social Survey

ifit: Data: ifit Intensive Longitudinal Data

iri: Data: Interpersonal Reactivity Index (IRI)

ptsd: Data: Post-Traumatic Stress Disorder

ptsd_cor1: Data: Post-Traumatic Stress Disorder (Sample # 1)

ptsd_cor2: Data: Post-Traumatic Stress Disorder (Sample # 2)

ptsd_cor3: Data: Post-Traumatic Stress Disorder (Sample # 3)

ptsd_cor4: Data: Post-Traumatic Stress Disorder (Sample # 4)

rsa: Data: Resilience Scale of Adults (RSA)

Sachs: Data: Sachs Network

tas: Data: Toronto Alexithymia Scale (TAS)

women_math: Data: Women and Mathematics