3.5.3.3.15. statsmodels.genmod.generalized_estimating_equations.ParameterConstraint¶
-
class
statsmodels.genmod.generalized_estimating_equations.
ParameterConstraint
(lhs, rhs, exog)[source]¶ A class for managing linear equality constraints for a parameter vector.
Parameters: lhs : ndarray
A q x p matrix which is the left hand side of the constraint lhs * param = rhs. The number of constraints is q >= 1 and p is the dimension of the parameter vector.
rhs : ndarray
A 1-dimensional vector of length q which is the right hand side of the constraint equation.
exog : ndarray
The n x p exognenous data for the full model.
-
__init__
(lhs, rhs, exog)[source]¶ Parameters: lhs : ndarray
A q x p matrix which is the left hand side of the constraint lhs * param = rhs. The number of constraints is q >= 1 and p is the dimension of the parameter vector.
rhs : ndarray
A 1-dimensional vector of length q which is the right hand side of the constraint equation.
exog : ndarray
The n x p exognenous data for the full model.
-
3.5.3.3.15.1. Methods¶
__init__ (lhs, rhs, exog) |
|
||
offset_increment () |
Returns a vector that should be added to the offset vector to accommodate the constraint. | ||
reduced_exog () |
Returns a linearly transformed exog matrix whose columns span the constrained model space. | ||
restore_exog () |
Returns the full exog matrix before it was reduced to satisfy the constraint. | ||
unpack_cov (bcov) |
Converts the covariance matrix bcov from reduced to full coordinates. | ||
unpack_param (params) |
Converts the parameter vector params from reduced to full coordinates. |