Hi there
Please note that this repository is work in progress 🔨
- In the introduction, indicate what is the goal of this repository
- Include all R packages used
- Create a test to check for newer versions of packages and bonus to have it in a CI
- Normal vs Poisson
- Quadratic trend components for Poisson model
- Quadratic trend components for hidden Markov chains
- Softmax regressions
- Four latent classes
Placeholder
- 📄 ModelSpecifications
- 📁 ModelImplementations
- 📁 SimulationStudyData
- 📁 SimulationStudyResults
Define and implement a strategy for setting the hyperparameters listed below so that label switching is prevented. However, the hyperparameters are not allowed to be informative regarding classes.
- SD hyperparameters for Normal prior of constants
- SD hyperparameters for Normal prior of linear trend components
- SD hyperparameters for Normal prior of SDs for Normal distributions of dependent variable
- Except for the implementations of the GMMs, all operations have been performed using R (R Core Team, 2022)
- The GMMs have been implemented using Stan and estimated using the No-U-Turn Sampler (or NUTS for short) via RStan: the R interface to Stan (Hoffman & Gelman, 2014; Stan Development Team, n.d.; Stan Development Team, 2024)
- In the context of parameter initializations for the NUTS, the R Stats Package has been used to apply Hartigan and Wong's (1979) K-means clustering algorithm with maximum ten iterations and two random sets (R Core Team and contributors worldwide, 2022)
- Moreover, the following R packages have been used: package by author (year), ..., and ...
- R Core Team. (2022). R: A Language and Environment for Statistical Computing (Version 4.2.2) [Programming language]. The R Project for Statistical Computing.
- R Core Team and contributors worldwide. (2022). The R Stats Package (Version 4.2.2) [R package]. The R Project for Statistical Computing.
- Stan Development Team. (2024). RStan: the R interface to Stan (Version 2.32.6) [R package]. The R Project for Statistical Computing.
- Hartigan, J. A. and Wong, M. A. (1979). Algorithm AS 136: A K-means clustering algorithm. Applied Statistics, 28(1), 100-108.
- Hoffman, M. D. and Gelman, A. (2014). The No-U-Turn Sampler: Adaptively Setting Path Lengths in Hamiltonian Monte Carlo. Journal of Machine Learning Research, 15, 1593-1623.
- Stan Development Team. (n.d.). Stan Documentation Version 2.34. Stan. https://mc-stan.org/docs/