Skip to content

Feat/zinb simulation and toy model#6

Open
Causal-Inference wants to merge 2 commits intomacork:mainfrom
Causalitat:feat/zinb-simulation-and-toy-model
Open

Feat/zinb simulation and toy model#6
Causal-Inference wants to merge 2 commits intomacork:mainfrom
Causalitat:feat/zinb-simulation-and-toy-model

Conversation

@Causal-Inference
Copy link
Copy Markdown

No description provided.

Implements several enhancements to the simulation framework:

1.  Introduces a Zero-Inflated Negative Binomial (ZINB) distribution
    as a new option for treatment generation (gps_mod = 5) in
    `functions/simulation_functions.R`. This allows for simulating
    count-based treatments with excess zeros. Includes parameters
    zinb_mu, zinb_theta, and zinb_pi for customization.

2.  Adds `toy_model_simulation.R`, a simplified and heavily commented
    script. This serves as an accessible entry point for users to
    understand the simulation mechanics by running a single,
    configurable scenario.

3.  Updates `1_run_simulation.R` to include the new ZINB gps_mod
    in its iteration loop.

4.  Enhances documentation in `README.md` to describe the toy model
    and the new ZINB treatment generation mechanism. Code comments
    were also added/updated for clarity.
Introduces `nb_model_simulation.R`, a standalone script demonstrating
how to simulate Negative Binomial (NB) distributed treatment.

This script utilizes the existing ZINB generation mechanism (gps_mod = 5)
by setting the zero-inflation probability (zinb_pi) to 0.
It is structured similarly to toy_model_simulation.R for clarity
and ease of use, serving as a clear example for users wanting to
simulate NB treatments.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant