emos_model: Fit EMOS coefficients a, b_k, c, d for truncated normal b_k,...
In kdayday/forecasting: Probabilistic Solar Forecsting

Fit EMOS coefficients a, b_k, c, d for truncated normal b_k, c, and d are parameterized as squares to control non-negativity of regression and scale parameters, like the "square" coefRule and varRule in the ensembleMOS package.

1	emos_model(tel, ens, max_power, par_init = NA)

`tel`	Vector of training telemetry data
`ens`	Matrix of training ensemble member data [time x member]
`max_power`	site AC rating, for upper limit of truncated normal
`par_init`	Initial parameter values as a list with a, b, c, d