WeightedSumFitter

class causalpy.pymc_models.WeightedSumFitter

Used for synthetic control experiments.

Defines the PyMC model:

\[\begin{split}\sigma &\sim \mathrm{HalfNormal}(1) \\ \beta &\sim \mathrm{Dirichlet}(1,...,1) \\ \mu &= X \cdot \beta \\ y &\sim \mathrm{Normal}(\mu, \sigma) \\\end{split}\]

Example

>>> import causalpy as cp
>>> import numpy as np
>>> import xarray as xr
>>> from causalpy.pymc_models import WeightedSumFitter
>>> sc = cp.load_data("sc")
>>> control_units = ['a', 'b', 'c', 'd', 'e', 'f', 'g']
>>> X = xr.DataArray(
...     sc[control_units].values,
...     dims=["obs_ind", "coeffs"],
...     coords={"obs_ind": sc.index, "coeffs": control_units},
... )
>>> y = xr.DataArray(
...     sc['actual'].values.reshape((sc.shape[0], 1)),
...     dims=["obs_ind", "treated_units"],
...     coords={"obs_ind": sc.index, "treated_units": ["actual"]},
... )
>>> coords = {
...     "coeffs": control_units,
...     "treated_units": ["actual"],
...     "obs_ind": np.arange(sc.shape[0]),
... }
>>> wsf = WeightedSumFitter(sample_kwargs={"progressbar": False})
>>> wsf.fit(X, y, coords=coords)
Inference data...

Methods

WeightedSumFitter.__init__([sample_kwargs])
WeightedSumFitter.add_coord(name[, values, ...])	Register a dimension coordinate with the model.
WeightedSumFitter.add_coords(coords, *[, ...])	Vectorized version of Model.add_coord.
WeightedSumFitter.add_named_variable(var[, dims])	Add a random graph variable to the named variables of the model.
WeightedSumFitter.build_model(X, y, coords)	Defines the PyMC model
WeightedSumFitter.calculate_cumulative_impact(impact)
WeightedSumFitter.calculate_impact(y_true, ...)
WeightedSumFitter.check_start_vals(start, ...)	Check that the logp is defined and finite at the starting point.
WeightedSumFitter.compile_d2logp([vars, ...])	Compiled log probability density hessian function.
WeightedSumFitter.compile_dlogp([vars, jacobian])	Compiled log probability density gradient function.
WeightedSumFitter.compile_fn(outs, *[, ...])	Compiles a PyTensor function.
WeightedSumFitter.compile_logp([vars, ...])	Compiled log probability density function.
WeightedSumFitter.copy()	Clone the model.
WeightedSumFitter.create_value_var(rv_var, ...)	Create a TensorVariable that will be used as the random variable's "value" in log-likelihood graphs.
WeightedSumFitter.d2logp([vars, jacobian, ...])	Hessian of the models log-probability w.r.t.
WeightedSumFitter.debug([point, fn, verbose])	Debug model function at point.
WeightedSumFitter.dlogp([vars, jacobian])	Gradient of the models log-probability w.r.t.
WeightedSumFitter.eval_rv_shapes()	Evaluate shapes of untransformed AND transformed free variables.
WeightedSumFitter.fit(X, y[, coords])	Draw samples from posterior, prior predictive, and posterior predictive distributions, placing them in the model's idata attribute.
WeightedSumFitter.get_context([...])
WeightedSumFitter.initial_point([random_seed])	Compute the initial point of the model.
WeightedSumFitter.logp([vars, jacobian, sum])	Elemwise log-probability of the model.
WeightedSumFitter.logp_dlogp_function([...])	Compile a PyTensor function that computes logp and gradient.
WeightedSumFitter.make_obs_var(rv_var, data, ...)	Create a TensorVariable for an observed random variable.
WeightedSumFitter.name_for(name)	Check if name has prefix and adds if needed.
WeightedSumFitter.name_of(name)	Check if name has prefix and deletes if needed.
WeightedSumFitter.point_logps([point, ...])	Compute the log probability of point for all random variables in the model.
WeightedSumFitter.predict(X)	Predict data given input data X
WeightedSumFitter.print_coefficients(labels)
WeightedSumFitter.profile(outs, *[, n, ...])	Compile and profile a PyTensor function which returns outs and takes values of model vars as a dict as an argument.
WeightedSumFitter.register_data_var(data[, dims])	Register a data variable with the model.
WeightedSumFitter.register_rv(rv_var, name, *)	Register an (un)observed random variable with the model.
WeightedSumFitter.replace_rvs_by_values(...)	Clone and replace random variables in graphs with their value variables.
WeightedSumFitter.score(X, y)	Score the Bayesian \(R^2\) given inputs X and outputs y.
WeightedSumFitter.set_data(name, values[, ...])	Change the values of a data variable in the model.
WeightedSumFitter.set_dim(name, new_length)	Update a mutable dimension.
WeightedSumFitter.set_initval(rv_var, initval)	Set an initial value (strategy) for a random variable.
WeightedSumFitter.shape_from_dims(dims)
WeightedSumFitter.to_graphviz(*[, ...])	Produce a graphviz Digraph from a PyMC model.

Attributes

basic_RVs	List of random variables the model is defined in terms of.
continuous_value_vars	All the continuous value variables in the model.
coords	Coordinate values for model dimensions.
datalogp	PyTensor scalar of log-probability of the observed variables and potential terms.
dim_lengths	The symbolic lengths of dimensions in the model.
discrete_value_vars	All the discrete value variables in the model.
isroot
observedlogp	PyTensor scalar of log-probability of the observed variables.
parent
potentiallogp	PyTensor scalar of log-probability of the Potential terms.
prefix
root
unobserved_RVs	List of all random variables, including deterministic ones.
unobserved_value_vars	List of all random variables (including untransformed projections), as well as deterministics used as inputs and outputs of the model's log-likelihood graph.
value_vars	List of unobserved random variables used as inputs to the model's log-likelihood (which excludes deterministics).
varlogp	PyTensor scalar of log-probability of the unobserved random variables (excluding deterministic).
varlogp_nojac	PyTensor scalar of log-probability of the unobserved random variables (excluding deterministic) without jacobian term.

__init__(sample_kwargs=None)

Parameters:

sample_kwargs (Optional[Dict[str, Any]]) – A dictionary of kwargs that get unpacked and passed to the pymc.sample() function. Defaults to an empty dictionary.

classmethod __new__(*args, **kwargs)