Information
- MIC (Maximum Information Coefficient) is a measure of the strength of the linear or nonlinear association between two variables.
- MIC ranges between 0 and 1, where 0 indicates no association and 1 indicates perfect association.
- It is a nonparametric method that uses mutual information as a measure of association between two variables.
- MIC is scale-invariant, which means it can be used to compare the association between variables measured in different units or scales.
- It can be used for both continuous and discrete data and is robust to outliers.
- MIC has been shown to outperform other popular correlation measures like Pearson's correlation and Spearman's rank correlation in certain scenarios.
'Zettelkasten > Terminology Information' 카테고리의 다른 글
| ReLU (Rectified Linear Unit) (0) | 2023.04.19 |
|---|---|
| MLP (Multi-Layer Perceptron) (0) | 2023.04.18 |
| RES (Renewable Energy Source) (0) | 2023.04.17 |
| ESS (Energy Storage System) (0) | 2023.04.16 |
| BC Hydro (British Columbia Hydro) dataset (0) | 2023.04.16 |
댓글