Package index

The `Trial` class and run helpers

The run_trial() and run_trials() functions use the specified partner_selection, interaction, and iterate_model functions to specify these dynamics and iterate the model (an instance of socmod::AgentBasedModel). fixated helps us tell when all agents are performing the same behavior, i.e., the population has fixated.

run_trial(): Trial runner helper function

run_trials(): Run a grid of trial ensembles with parameter metadata

Trial: Trial class for running a single simulation

fixated(): Predicate function: has the model fixated on a single behavior?

Summarise a collection of trial outcomes or prevalence dynamics over model parameters

summarise_prevalence(): Summarize behavior prevalence over time within or across trials

summarise_outcomes(): Summarize outcomes across trials by input parameters

Visualizations

Tools to plot networks, outcomes, and time series

plot_prevalence(): Plot adoption counts of selected behaviors over time Plot adoption counts of selected behaviors (tracked_behaviors) over time.

plot_homophilynet(): Plot homophily network where groups color nodes.

plot_friendship_paradox(): Plot Friendship Paradox Comparison

ggnetplot(): Command that operates like ggplot

SOCMOD_PLOT_PALETTE: Custom color palette for scientific plots

Examples with real-world data

Simple examples of real-world data in socmod.

simulate_feld_1991(): Simulate Feld's 1991 Marketville Network

get_feld_1991_network(): Load Feld's 1991 data.

AgentBasedModel and ModelParameters

Create agent-based models and specify model configuration parameters

make_abm(): Create an AgentBasedModel instance

initialize_agents(): Initialize agents with adaptive and legacy behaviors

make_model_parameters(): Wrapper for initializing new ModelParameters instance.

AgentBasedModel: Agent-based model class

ModelParameters: Overwrite existing auxiliary parameters.

DEFAULT_PARAMETERS: Default parameters to create an agent-based model.

Learning strategy class, make_learning_strategy, and other tools

Functions to pass to run for model iteration in a learning model, and for either success-biased or frequency-biased adaptive learning.

iterate_learning_model(): A generic method for iterating a learning model, setting the current behavior and fitness to be whatever was identified as the next behavior and fitness.

LearningStrategy: LearningStrategy

success_bias_interact(): Success-biased interaction function

success_bias_learning_strategy: Define success-biased learning strategy.

success_bias_select_teacher(): Success-biased teacher selection

frequency_bias_interact(): Interaction function for frequency-biased adaptive learning.

frequency_bias_learning_strategy: Define frequency-biased learning strategy.

frequency_bias_select_teacher(): Frequency biased teacher selection does nothing

contagion_interaction(): Contagion-based interaction

contagion_learning_strategy: Define contagion learning "strategy".

contagion_model_step(): Contagion model step

contagion_partner_selection(): Contagion-based partner selection

dummy_learning_strategy(): Make a dummy learning strategy for mockups and testing.

make_learning_strategy(): Factory function for creating a LearningStrategy

learning_model_step(): More modern v0.1 alias for iterate_learning_model.

Agents and their neighbors

Agents in the model are represented by a number of attributes, with select class methods to track lists of their neighbors.

Agent: Agent class for socmod

Neighbors: Neighbors class

Networks

Create networks, including a growing suite of network construction routines.

make_homophily_network(): Create an undirected asymmetric homophily network.

make_small_world(): Make a small world network.

make_preferential_attachment(): Make a preferential attachment network.

G_NM(): Erdős-Rényi random graph G(N, M).

make_regular_lattice(): Create a regular lattice graph.

plot_homophilynet(): Plot homophily network where groups color nodes.

ggnetplot(): Command that operates like ggplot

add_unique_edge(): Add an undirected edge from v1 to v2 to graph g if it does not already exist.

get_all_possible_edges(): Get all possible edges between node indices 1 to N for either directed or undirected networks.

not_adjacent(): Check two vertices are not adjacent.

compare_friendship_paradox(): Compare Friendship Paradox in a Network

load_igraph_from_csv(): Load an Undirected igraph Object from a CSV Edge List

plot_friendship_paradox(): Plot Friendship Paradox Comparison