PTMLocScale.from_nparam()

PTMLocScale.from_nparam()#

classmethod PTMLocScale.from_nparam(y, nparam, knots_lo=-3.0, knots_hi=3.0, transition_width=0.3, normalization_tau2=None, scale_after_transformation=False, scaling_factor=None, shape_param_prior=ShapePrior.RANDOM_WALK)[source]#

Initializes a penalized transformation model automatically from the desired number of shape parameters.

Parameters:

  • y (Any) – The observed response variable.

  • nparam (int) – The number of parameters in the spline segment of the transformation function. Called \(J\) in the paper.

  • knots_lo (float) – Lower and upper boundary knot, used to define an equidistant knot grid for the transformation function. (default: -3.0)

  • knots_hi (float) – Lower and upper boundary knot, used to define an equidistant knot grid for the transformation function. (default: 3.0)

  • transition_width (float) – The width of the transition segment of the transformation function. A number larger than 0, indicating the length in terms of a multiple of the range of interior knots. Called \(\lambda\). (default: 0.3)

  • scale_after_transformation (bool) – Whether to scale the transformed residuals. (default: False)

  • scaling_factor (Optional[Var]) – The scaling factor \(\omega\) for the transformation function. If None, the scaling factor will be fixed to one. (default: None)

  • shape_param_prior (ShapePrior) – The prior distribution for the shape parameter. (default: <ShapePrior.RANDOM_WALK: 1>)

Return type:

PTMLocScale

Contents