Class of a loess fit outputs
This object is automatically created with empty values when a
new loess object is instantiated. The object gets filled when the
loess.fit() method is called.
- Parameters:
- n
int Number of independent observation
- p
int Number of variables
-
diagonal
ndarray - Diagonal of the operator hat matrix,
shape (n,)
-
divisor
ndarray - Normalization divisors for numeric
predictors, shape (p,)
-
enp
float - Equivalent number of parameters
-
fitted_residuals
ndarray - Fitted residuals, shape (n,),
(observations - fitted values)
-
fitted_values
ndarray - Fitted values, shape (n,)
-
one_delta
float - Statistical parameter used in the computation
of standard errors
-
pseudovalues
ndarray - Adjusted values of the response
when robust estimation is used, shape (n,)
-
residual_scale
float - Estimate of the scale of residuals
-
robust
ndarray - Robustness weights for robust fitting,
shape (n,)
-
trace_hat
float - Trace of the operator hat matrix
-
two_delta
float - Statistical parameter used in the computation
of standard errors