Uncertainty and Negative Word-of-Mouth Model

Version: Aug 2025

This repository contains the Julia implementation of a model used to study the role of consumer uncertainty for NWoM’s impact on adoption dynamics.

Getting Started

These instructions will allow you to run the model on your system.

System Requirements and Installation

To run this code, you need to install Julia (version 1.11.2 recommended).

The following Julia packages are required (specific versions recommended for compatibility):

• Agents - Version 6.1.10
• ArgParse - Version 1.2.0
• CSV - Version 0.10.15
• DataFrames - Version 1.7.0
• DataStructures - Version 0.18.20
• Distributions - Version 0.25.112
• Plots - Version 1.40.8
• QuadGK - Version 2.11.1
• StatsBase - Version 0.34.3
• StatsPlots - Version 0.15.7
• TensorCast - Version 0.4.8

To install the required packages, start Julia and run:

using Pkg; Pkg.add("<package name>")

Alternatively, if the Project.toml is included, you can install all dependencies exactly as used in development:

using Pkg
Pkg.instantiate()

Model Structure and Execution

The core implementation of the model is located in the src/model/ directory. To run the model, an initial state must be defined. The baseline initialization is provided in src/config/init_monopoly.jl. The subset of data collected during a simulation run is configured via src/config/data_collection_monopoly.jl.

To run an experiment (i.e. multiple parallel simulation batches), execute run_exp.jl. Each experiment must be defined in a configuration file - see experiments/monopoly_goldenberg.jl for an example. Use the following command:

julia -p <no_cpus> run_exp.jl <name-of-config-file> [--chunk <i>] [--no_chunks <n>]

• -p <no_cpus> : number of CPU cores to use in parallel
• --chunk and --no_chunk: optional parameters to split the experiment into chunks, useful for distributed execution across multiple machines

Replication

To reproduce the results from the paper by re-simulating the model, use the following commands:

julia -p <no_cpus> run_exp.jl monopoly_goldenberg
julia -p <no_cpus> run_exp.jl monopoly_bayes
julia -p <no_cpus> run_exp.jl monopoly_cosita
julia -p <no_cpus> run_exp.jl monopoly_ws
julia -p <no_cpus> run_exp.jl monopoly_ba
julia -p <no_cpus> run_exp.jl monopoly_er
julia -p <no_cpus> run_exp.jl monopoly_mixed_cosita_bayes
julia -p <no_cpus> run_exp.jl monopoly_mixed_cosita_goldenberg
julia -p <no_cpus> run_exp.jl monopoly_mixed_emp

The resulting data will be stored in data/, which by default contains the data used to create the plots in the paper.

In order to recreate all plots from the paper, run:

julia plots_paper.jl

Empirical Data

We conducted a survey among German consumers. The original survey data (in xlsx file format) can be found in the data/ folder.

Authors

Herbert Dawid, Dirk Kohlweyer, Melina Schleef, Christian Stummer, Frederik Tolkmitt

Further Links

• ETACE - Economic Theory and Computational Economics, Bielefeld University
• ITM - Innovation and Technology Management, Bielefeld University