.rst .pdf

repository open issue

larch.model.model_group.ModelGroup

class ModelGroup(models, *, parameters=None, frame=None, title=None, dataservice=None, constraints=None)

Bases: larch.model.abstract_model.AbstractChoiceModel, collections.abc.MutableSequence

__init__(models, *, parameters=None, frame=None, title=None, dataservice=None, constraints=None)

Methods

__init__(models, *[, parameters, frame, ...])
append(value)	S.append(value) -- append value to the end of the sequence
batch_update(self)	This context manager suppresses internal updates when editing parameters.
bhhh
calculate_parameter_covariance	Compute the parameter covariance matrix.
check_d_loglike	Check that the analytic and finite-difference gradients are approximately equal.
clear()
clear_best_loglike
count(value)
cross_validate	A simple but well optimized cross-validated log likelihood.
d2_loglike	Compute the (approximate) second derivative of log likelihood with respect to the parameters.
d_loglike	Compute the first derivative of log likelihood with respect to the parameters.
d_probability	Compute the partial derivative of probability w.r.t.
doctor([repair_ch_av, repair_ch_zq, ...])
estimate	A convenience method to load data, maximize loglike, and get covariance.
estimation_statistics	Create an XHTML summary of estimation statistics.
extend(values)	S.extend(iterable) -- extend sequence by appending elements from the iterable
get_holdfast(self, name, *[, default])
get_slot_x(self, name[, holdfast_invalidates])	Get the position of a named parameter within the parameters index.
get_value(self, name, *[, default])
get_values(self)
index(value, [start, [stop]])	Raises ValueError if the value is not present.
insert(i, value)	S.insert(index, value) -- insert value before index
jumpstart_bhhh	Jump start optimization
likelihood_ratio	Compute a likelihood ratio for changing a parameter to its null value.
load_data([dataservice, autoscale_weights, ...])
lock_value(self, name, value[, note, ...])	Set a fixed value for a model parameter.
lock_values(self, *names[, note])	Set a fixed value for one or more model parameters.
loglike([x, start_case, stop_case, ...])	Compute a log likelihood value.
loglike2([x, start_case, stop_case, ...])	Compute a log likelihood value and its first derivative.
loglike2_bhhh	Compute a log like, it first deriv, and the BHHH approx of the Hessian.
loglike3	Compute a log likelihood value, it first derivative, and the finite-difference approximation of the Hessian.
loglike_constants_only
loglike_nil	Compute the log likelihood with no model at all.
loglike_null	Compute the log likelihood at null values.
mangle(self, *args, **kwargs)
maximize_loglike	Maximize the log likelihood.
neg_d_loglike
neg_loglike
neg_loglike2
neg_loglike3
noop
parameter_summary(self[, output])	Create a tabular summary of parameter values.
pf_sort(self)	Sort (on index, i.e. parameter name) and return the parameter frame.
pfo(self)
pformat(self, parameter_name[, ...])	Get the value of a parameter or a parameter expression.
pop([index])	Raise IndexError if list is empty or index is out of range.
pvalue(self, parameter_name[, ...])	Get the value of a parameter or a parameter expression.
remove(value)	S.remove(value) -- remove first occurrence of value.
reverse()	S.reverse() -- reverse IN PLACE
rho_sq_nil	Compute the rho squared value w.r.t.
rho_sq_null	Compute the rho squared value w.r.t.
set_cap(self[, cap])	Set the parameter values for one or more parameters.
set_frame(self, frame)	Assign a new parameter frame.
set_value(self, name[, value])	Set the value for a single model parameter.
set_values([values, respect_holdfast])	Set the parameter values for one or more parameters.
shift_values(self, shifts)
simple_fit_bhhh	Makes a series of steps using the BHHH algorithm.
simple_step_bhhh	Makes one step using the BHHH algorithm.
to_xlsx(filename[, save_now])
total_weight()	The total weight of cases in the attached data of all grouped models.
unmangle(self[, force])
update_values(self, values)

Attributes

constrained_covariance_matrix
constraints
covariance_matrix
dataframes
hessian_matrix
most_recent_estimation_result
n_cases	Total number of cases in the attached data of all grouped models.
ordering
ordering_tail
pbounds	A copy of the current min-max bounds of the parameters.
pf	The parameter frame, unmangling on access.
pnames	A copy of the current names of the parameters.
possible_overspecification
pvals	A copy of the current value of the parameters.
robust_covariance_matrix
title

previous

ModelGroup

next

larch.model.model_group.ModelGroup.n_cases