yeojohnson_llf¶
- mpsci.stats.yeojohnson_llf(lam, x)¶
The log-likelihood function for the Yeo-Johnson transformation.
- Parameters:
lam (scalar) – Parameter for Yeo-Johnson transformation. See yeojohnson for details.
x (sequence of numbers) – Data to calculate Yeo-Johnson log-likelihood for.
- Returns:
llf – Yeo-Johnson log-likelihood of x given lam.
- Return type:
float
Notes
The Yeo-Johnson log-likelihood is
\[\frac{-N}{2} \log\left(\hat{\sigma}^2\right) + (\lambda - 1)\sum_{i}\textrm{sign}(x_i) \log(|x_i| + 1)\]where \(N\) is
len(x)and \(\hat{\sigma}^2\) is the estimated variance of the Yeo-Johnson transformed input datax.