This code reproduces the experiments in "Selective inference for clustering with unknown variance" (authors: Youngjoo Yun and Rina Foygel Barber), available at http://arxiv.org/abs/2301.12999.
Each of the files starting with "fig_" corresponds to a figure in the paper, and the correspondence is shown below. Running each of these files saves the same figure into a folder called "plots" and saves the workspace into a folder called "workspaces" in the working directory. Running fig_penguins.R further saves the results in [Table 1] into a folder called "results."
- [Figure 1] fig_motivation.R
- [Figure 2] fig_decomposition.R
- [Figure 3] fig_type1.R
- [Figure 4] fig_power_K2.R, fig_power_K3_equidistant.R, and fig_power_K3_horizontal.R
- [Figure 5] fig_type1_t5.R
- [Figure 6] fig_type1_t10.R
- [Figure 7] fig_type1_noniso.R
- [Figure 8] fig_penguins.R
The functions used in the scripts above are stored in the file called functions.R.