https://github.com/santikka/sima
Simulation framework for realistic large-scale individual-level health data generation
https://github.com/santikka/sima
healthcare-application population simulation simulation-framework simulator synthetic-data
Last synced: 2 months ago
JSON representation
Simulation framework for realistic large-scale individual-level health data generation
- Host: GitHub
- URL: https://github.com/santikka/sima
- Owner: santikka
- Created: 2020-06-09T08:53:03.000Z (about 5 years ago)
- Default Branch: master
- Last Pushed: 2021-06-05T07:27:50.000Z (about 4 years ago)
- Last Synced: 2023-12-16T15:16:31.587Z (over 1 year ago)
- Topics: healthcare-application, population, simulation, simulation-framework, simulator, synthetic-data
- Language: R
- Homepage: https://santikka.github.io/Sima/
- Size: 184 KB
- Stars: 2
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.Rmd
Awesome Lists containing this project
README
---
output: github_document
---
```{r, echo = FALSE}
knitr::opts_chunk$set(
collapse = TRUE,
comment = "#>"
)
```
# Simulation Framework
A simulation framework for large-scale data generation with applications in the health care domain. The work is a part of the DEMO project http://www.jyu.fi/demo.
## Installation
Install the released version of the framework from CRAN (not released yet):
```{r, eval = FALSE}
install.packages("Sima")
```
Alternatively, install the development version from GitHub:
```{r, eval = FALSE}
# install.packages("devtools")
devtools::install_github("santikka/Sima")
```
## Usage
A simulator consists of events and an initial population. Events come in two types of R6 class objects, known as manipulation events and accumulation events. Manipulation events modify the status of initial population whereas accumulation events introduce new individuals into the population. The population itself is contained in a `data.table` and can be either entirely synthetic or based on real data. To construct a simulator, first define the events governing the system of interest and the initial population.
Read `vignette("Sima")` for more details on how to construct a simulator for your purposes.
## Recent changes
### Changes from version 0.5.1 to 0.5.2
* The package no longer depends on 'doMPI', 'foreach' and 'iterators' packages. Now suggests instead.
* Added explicit support for interventions via the method 'intervene'.
* All event parameters are now named.
* Now supports derived parameters for events, that change their values automatically if the parameter values change.
* Events can now have their own persistent data.
* Removed `Time` class as unnecessary.