EmpiricalCovariance
The maximum-likelihood (sample) covariance estimator.
Algorithm
Centres the data (unless assume_centered) and forms the 1/n scatter matrix. The maximum-likelihood normalisation matches scikit-learn.
Constructor
Skigen::EmpiricalCovariance<Scalar> model(bool assume_centered = false);
Parameters
| Parameter | Default | Description |
|---|---|---|
assume_centered | false | Skip mean subtraction if data is already centred. |
Methods
| Method | Description |
|---|---|
fit(X) | Estimate the covariance. |
score(X) | Gaussian log-likelihood of new data. |
Fitted Attributes
| Accessor | Description |
|---|---|
covariance() | Estimated p×p covariance. |
location() | Estimated mean. |
Example
Skigen::EmpiricalCovariance<double> cov;
cov.fit(X);
auto C = cov.covariance();
Verified against scikit-learn
This estimator is checked by the parity suite. See the generator tests/parity/generate_covariance_reference.py and the reference fixtures in tests/parity/data/empirical_covariance/, exercised by tests/parity/parity_covariance.cpp.
API Reference
For full signatures see the EmpiricalCovariance API Reference.