calculate standard deviation and confidence interval for prediction
applies to WLS and OLS, not to general GLS, that is independently but not identically distributed observations
Parameters : | res : regression result instance
exog : array_like (optional)
weights : scalar or array_like (optional)
alpha : float (default: alpha = 0.5)
|
---|---|
Returns : | predstd : array_like, 1d
interval_l, interval_u : array_like
|
Notes
The result instance needs to have at least the following res.model.predict() : predicted values or res.fittedvalues : values used in estimation res.cov_params() : covariance matrix of parameter estimates
If exog is 1d, then it is interpreted as one observation, i.e. a row vector.
testing status: not compared with other packages
References
Greene p.111 for OLS, extended to WLS by analogy