def summary(self):
"""Summary statistics describing the fit.
Returns
-------
df : DataFrame
Contains columns coef, np.exp(coef), se(coef), z, p, lower, upper"""
ci = 100 * (1 - self.alpha)
z = inv_normal_cdf(1 - self.alpha / 2)
with np.errstate(invalid="ignore",
divide="ignore",
over="ignore",
under="ignore"):
df = pd.DataFrame(index=self.params_.index)
df["coef"] = self.params_
df["exp(coef)"] = self.hazard_ratios_
df["se(coef)"] = self.standard_errors_
df["coef lower %g%%" %
ci] = self.confidence_intervals_["%g%% lower-bound" % ci]
df["coef upper %g%%" %
ci] = self.confidence_intervals_["%g%% upper-bound" % ci]
df["exp(coef) lower %g%%" %
ci] = self.hazard_ratios_ * np.exp(-z * self.standard_errors_)
df["exp(coef) upper %g%%" %
ci] = self.hazard_ratios_ * np.exp(z * self.standard_errors_)
df["z"] = self._compute_z_values()
df["p"] = self._compute_p_values()
df["-log2(p)"] = -utils.quiet_log2(df["p"])
return df
def print_summary(self, decimals=2, style=None, columns=None, **kwargs):
"""
Print summary statistics describing the fit, the coefficients, and the error bounds.
Parameters
-----------
decimals: int, optional (default=2)
specify the number of decimal places to show
style: string
{html, ascii, latex}
columns:
only display a subset of ``summary`` columns. Default all.
kwargs:
print additional meta data in the output (useful to provide model names, dataset names, etc.) when comparing
multiple outputs.
"""
justify = string_rjustify(18)
headers = []
if self.event_col:
headers.append(("event col", "'%s'" % self.event_col))
if self.weights_col:
headers.append(("weights col", "'%s'" % self.weights_col))
if isinstance(self.penalizer, np.ndarray) or self.penalizer > 0:
headers.append(("penalizer", self.penalizer))
if self.strata:
headers.append(("strata", self.strata))
headers.extend([
("number of subjects", self._n_unique),
("number of periods", self._n_examples),
("number of events", self.event_observed.sum()),
("partial log-likelihood",
"{:.{prec}f}".format(self.log_likelihood_, prec=decimals)),
("time fit was run", self._time_fit_was_called),
])
sr = self.log_likelihood_ratio_test()
footers = []
footers.extend([
("Partial AIC", "{:.{prec}f}".format(self.AIC_partial_,
prec=decimals)),
(
"log-likelihood ratio test",
"{:.{prec}f} on {} df".format(sr.test_statistic,
sr.degrees_freedom,
prec=decimals),
),
("-log2(p) of ll-ratio test",
"{:.{prec}f}".format(-utils.quiet_log2(sr.p_value),
prec=decimals)),
])
p = Printer(self, headers, footers, justify, kwargs, decimals, columns)
p.print(style=style)
def to_ascii(self, decimals=2, **kwargs):
extra_kwargs = dict(list(self._kwargs.items()) + list(kwargs.items()))
meta_data = self._stringify_meta_data(extra_kwargs)
df = self.summary
df["-log2(p)"] = -utils.quiet_log2(df["p"])
s = self.__repr__()
s += "\n" + meta_data + "\n"
s += "---\n"
s += df.to_string(float_format=format_floats(decimals),
index=self.name is not None,
formatters={"p": format_p_value(decimals)})
return s
def summary(self):
"""
Returns
-------
DataFrame
a DataFrame containing the test statistics and the p-value
"""
cols = ["test_statistic", "p"]
# test to see if self.names is a tuple
if self.name and isinstance(self.name[0], tuple):
index = pd.MultiIndex.from_tuples(self.name)
else:
index = self.name
df = pd.DataFrame(list(zip(self._test_statistic, self._p_value)), columns=cols, index=index).sort_index()
df["-log2(p)"] = -utils.quiet_log2(df["p"])
return df
def deviance(self):
return -utils.quiet_log2(self.log_likelihood_ratio_test().p_value)