Upgrading statsmodels 0.11 to 0.12
Unknown0.11.1 → 0.12.2New8Fixed2
Fetching 4 releases from GitHub…3 / 4
UnknownUpgrade verdict
DepDelta is still reading the notes for 1 of 4 releases, so breaking changes can't be ruled out yet.
In plain terms
- DepDelta is still reading the release notes in this range, so it can't rule out breaking changes yet.
- This page updates as the notes come in.
Must act Still reading release notes
DepDelta is still reading the release notes in this range. This updates as they come in.
Python requirement changed in 0.12.0rc0: >=3.5 → >=3.6.
API changes
Generated by DepDelta from published Python source. Not mixed with prose notes.
- removed:
export iteritems - removed:
export iterkeys - removed:
export itervalues
- added:
function to_numpy(po: pd.DataFrame) -> np.ndarray - added:
variable pandas_lt_1_0_0
- removed:
function iteritems(obj, **kwargs) - removed:
function iterkeys(obj, **kwargs) - removed:
function itervalues(obj, **kwargs)
- added:
variable SP_LT_15 - added:
variable SP_VERSION
- changed:
constructor CountModel(endog, exog, offset=…, exposure=…, missing=…, **kwargs) → constructor CountModel(endog, exog, offset=…, exposure=…, missing=…, check_rank=…, **kwargs) - changed:
constructor GeneralizedPoisson(endog, exog, p=…, offset=…, exposure=…, missing=…, **kwargs) → constructor GeneralizedPoisson(endog, exog, p=…, offset=…, exposure=…, missing=…, check_rank=…, **kwargs) - changed:
constructor MNLogit(endog, exog, **kwargs) → constructor MNLogit(endog, exog, check_rank=…, **kwargs) - changed:
constructor NegativeBinomial(endog, exog, loglike_method=…, offset=…, exposure=…, missing=…, **kwargs) → constructor NegativeBinomial(endog, exog, loglike_method=…, offset=…, exposure=…, missing=…, check_rank=…, **kwargs) - changed:
constructor NegativeBinomialP(endog, exog, p=…, offset=…, exposure=…, missing=…, **kwargs) → constructor NegativeBinomialP(endog, exog, p=…, offset=…, exposure=…, missing=…, check_rank=…, **kwargs)
- changed:
method PHReg.get_distribution(params) → method PHReg.get_distribution(params, scale=…, exog=…) - changed:
method PHReg.predict(params, exog=…, cov_params=…, endog=…, strata=…, offset=…, pred_type=…) → method PHReg.predict(params, exog=…, cov_params=…, endog=…, strata=…, offset=…, pred_type=…, pred_only=…) - changed:
method rv_discrete_float.rvs() → method rv_discrete_float.rvs(n=…)
- changed:
constructor Autoregressive(dist_func=…) → constructor Autoregressive(dist_func=…, grid=…)
- added:
method cauchy.inverse_deriv2(z) - added:
method probit.deriv2(p) - added:
method probit.inverse_deriv2(z)
- removed:
method GEE.predict(params, exog=…, offset=…, exposure=…, linear=…) - removed:
method GEEResults.fittedvalues() - removed:
method GEEResults.resid_anscombe() - removed:
method GEEResults.resid_deviance() - removed:
method GEEResults.resid_pearson() - removed:
method GEEResults.resid_response() - removed:
method GEEResults.resid_working() - changed:
class GEE(base.Model) → class GEE(GLM) - changed:
class GEEResults(base.LikelihoodModelResults) → class GEEResults(GLMResults)
- changed:
function qqline(ax, line, x=…, y=…, dist=…, fmt=…) → function qqline(ax, line, x=…, y=…, dist=…, fmt=…, **lineoptions)
- changed:
function plot_acf(x, ax=…, lags=…, *, alpha=…, use_vlines=…, unbiased=…, fft=…, missing=…, title=…, zero=…, vlines_kwargs=…, **kwargs) → function plot_acf(x, ax=…, lags=…, *, alpha=…, use_vlines=…, adjusted=…, fft=…, missing=…, title=…, zero=…, vlines_kwargs=…, **kwargs) - added:
function plot_predict(result, start=…, end=…, dynamic=…, alpha=…, ax=…, **predict_kwargs)
- changed:
method Summary.as_latex() → method Summary.as_latex(label=…)
- removed:
variable RANGE - removed:
variable a - removed:
variable b - removed:
variable bw - removed:
variable bw1 - removed:
variable bw2 - removed:
variable f - removed:
variable f2 - removed:
variable ft - removed:
variable grid - removed:
variable ifault - removed:
variable smooth - removed:
variable smooth1 - removed:
variable weights - removed:
variable xi - changed:
function kdensity(X, kernel=…, bw=…, weights=…, gridsize=…, adjust=…, clip=…, cut=…, retgrid=…) → function kdensity(x, kernel=…, bw=…, weights=…, gridsize=…, adjust=…, clip=…, cut=…, retgrid=…) - changed:
function kdensityfft(X, kernel=…, bw=…, weights=…, gridsize=…, adjust=…, clip=…, cut=…, retgrid=…) → function kdensityfft(x, kernel=…, bw=…, weights=…, gridsize=…, adjust=…, clip=…, cut=…, retgrid=…)
- changed:
function lowess(endog, exog, frac=…, it=…, delta=…, is_sorted=…, missing=…, return_sorted=…) → function lowess(endog, exog, frac=…, it=…, delta=…, xvals=…, is_sorted=…, missing=…, return_sorted=…)
- changed:
method MixedLM.fit(start_params=…, reml=…, niter_sa=…, do_cg=…, fe_pen=…, cov_pen=…, free=…, full_output=…, method=…, **kwargs) → method MixedLM.fit(start_params=…, reml=…, niter_sa=…, do_cg=…, fe_pen=…, cov_pen=…, free=…, full_output=…, method=…, **fit_kwargs) - changed:
method MixedLM.get_fe_params(cov_re, vcomp) → method MixedLM.get_fe_params(cov_re, vcomp, tol=…) - changed:
method MixedLMResults.profile_re(re_ix, vtype, num_low=…, dist_low=…, num_high=…, dist_high=…) → method MixedLMResults.profile_re(re_ix, vtype, num_low=…, dist_low=…, num_high=…, dist_high=…, **fit_kwargs)
- changed:
constructor RollingOLS(endog, exog, window=…, min_nobs=…, missing=…) → constructor RollingOLS(endog, exog, window=…, *, min_nobs=…, missing=…, expanding=…) - changed:
constructor RollingWLS(endog, exog, window=…, weights=…, min_nobs=…, missing=…) → constructor RollingWLS(endog, exog, window=…, *, weights=…, min_nobs=…, missing=…, expanding=…)
- added:
function iqr(a, c=…, axis=…) - added:
function qn(a, c=…, axis=…) - added:
function qn_scale(a, c=…, axis=…)
- removed:
export unitroot_adf
- added:
class HolderTuple(Holder) - added:
constructor HolderTuple(tuple_=…, **kwds)
- added:
function wald_test_noncent(params, r_matrix, value, results, diff=…, joint=…) - added:
function wald_test_noncent_generic(params, r_matrix, value, cov_params, diff=…, joint=…)
- removed:
class TestSimpleTable(object) - removed:
method TestSimpleTable.test_basic_1() - removed:
method TestSimpleTable.test_basic_1a() - removed:
method TestSimpleTable.test_basic_1b() - removed:
method TestSimpleTable.test_basic_2() - removed:
method TestSimpleTable.test_basic_2a() - removed:
method TestSimpleTable.test_basic_3() - removed:
method TestSimpleTable.test_basic_4() - removed:
method TestSimpleTable.test_basic_4a() - removed:
method TestSimpleTable.test_describe_summary_float_ndarray() - removed:
variable data1 - removed:
variable data2 - removed:
variable data3 - removed:
variable data4 - removed:
variable noperc - removed:
variable t1 - added:
class Description - added:
constructor Description(data: Union[np.ndarray, pd.Series, pd.DataFrame], stats: Sequence[str]=…, *, numeric: bool=…, categorical: bool=…, alpha: float=…, use_t: bool=…, percentiles: Sequence[Union[int, float]]=…, ntop: bool=…) - added:
field Description.categorical_statistics - added:
field Description.default_statistics - added:
field Description.numeric_statistics - added:
function describe(data: Union[np.ndarray, pd.Series, pd.DataFrame], stats: Sequence[str]=…, *, numeric: bool=…, categorical: bool=…, alpha: float=…, use_t: bool=…, percentiles: Sequence[Union[int, float]]=…, ntop: bool=…) -> pd.DataFrame - added:
function nancount(x, axis=…) - added:
function nankurtosis(arr, axis=…) - added:
function nanpercentile(arr, axis=…) - added:
function nanptp(arr, axis=…) - added:
function nanskewness(arr, axis=…) - added:
function nanuss(arr, axis=…) - added:
function pd_percentiles(df) - added:
function pd_ptp(df) - added:
method Description.categorical() -> pd.DataFrame - added:
method Description.frame() -> pd.DataFrame - added:
method Description.numeric() -> pd.DataFrame - added:
method Description.summary() -> SimpleTable - added:
variable CATEGORICAL_STATISTICS - added:
variable DEFAULT_STATISTICS - added:
variable DEPRECATION_MSG - added:
variable MISSING - added:
variable NUMERIC_STATISTICS - added:
variable PANDAS - added:
variable PERCENTILES - added:
variable QUANTILES - added:
variable ds
- changed:
function acorr_ljungbox(x, lags=…, boxpierce=…, model_df=…, period=…, return_df=…) → function acorr_ljungbox(x, lags=…, boxpierce=…, model_df=…, period=…, return_df=…, auto_lag=…) - changed:
function het_breuschpagan(resid, exog_het) → function het_breuschpagan(resid, exog_het, robust=…) - changed:
function linear_harvey_collier(res, order_by=…) → function linear_harvey_collier(res, order_by=…, skip=…)
- changed:
constructor Mediation(outcome_model, mediator_model, exposure, mediator=…, moderators=…, outcome_fit_kwargs=…, mediator_fit_kwargs=…) → constructor Mediation(outcome_model, mediator_model, exposure, mediator=…, moderators=…, outcome_fit_kwargs=…, mediator_fit_kwargs=…, outcome_predict_kwargs=…)
- added:
class CombineResults(object) - added:
constructor CombineResults(**kwds) - added:
function combine_effects(effect, variance, method_re=…, row_names=…, use_t=…, alpha=…, **kwds) - added:
function effectsize_2proportions(count1, nobs1, count2, nobs2, statistic=…, zero_correction=…, zero_kwds=…) - added:
function effectsize_smd(mean1, sd1, nobs1, mean2, sd2, nobs2) - added:
method CombineResults.conf_int(alpha=…, use_t=…) - added:
method CombineResults.conf_int_samples(alpha=…, use_t=…, nobs=…, ci_func=…) - added:
method CombineResults.plot_forest(ax=…, **kwds) - added:
method CombineResults.summary_array(alpha=…, use_t=…) - added:
method CombineResults.summary_frame(alpha=…, use_t=…) - added:
method CombineResults.test_homogeneity()
- changed:
function local_fdr(zscores, null_proportion=…, null_pdf=…, deg=…, nbins=…) → function local_fdr(zscores, null_proportion=…, null_pdf=…, deg=…, nbins=…, alpha=…)
- added:
function confint_mvmean(data, lin_transf=…, alpha=…, simult=…) - added:
function confint_mvmean_fromstats(mean, cov, nobs, lin_transf=…, alpha=…, simult=…) - added:
function test_cov(cov, nobs, cov_null) - added:
function test_cov_blockdiagonal(cov, nobs, block_len) - added:
function test_cov_diagonal(cov, nobs) - added:
function test_cov_oneway(cov_list, nobs_list) - added:
function test_cov_spherical(cov, nobs) - added:
function test_mvmean(data, mean_null=…, return_results=…) - added:
function test_mvmean_2indep(data1, data2)
- added:
function anova_generic(means, variances, nobs, use_var=…, welch_correction=…, info=…) - added:
function anova_oneway(data, groups=…, use_var=…, welch_correction=…, trim_frac=…) - added:
function confint_effectsize_oneway(f_stat, df, alpha=…, nobs=…) - added:
function confint_noncentrality(f_stat, df, alpha=…, alternative=…) - added:
function convert_effectsize_fsqu(f2=…, eta2=…) - added:
function effectsize_oneway(means, vars_, nobs, use_var=…, ddof_between=…) - added:
function equivalence_oneway(data, equiv_margin, groups=…, use_var=…, welch_correction=…, trim_frac=…, margin_type=…) - added:
function equivalence_oneway_generic(f_stat, n_groups, nobs, equiv_margin, df, alpha=…, margin_type=…) - added:
function equivalence_scale_oneway(data, equiv_margin, method=…, center=…, transform=…, trim_frac_mean=…, trim_frac_anova=…) - added:
function f2_to_wellek(f2, n_groups) - added:
function fstat_to_wellek(f_stat, n_groups, nobs_mean) - added:
function power_equivalence_oneway(f2_alt, equiv_margin, nobs_t, n_groups=…, df=…, alpha=…, margin_type=…) - added:
function simulate_power_equivalence_oneway(means, nobs, equiv_margin, vars_=…, k_mc=…, trim_frac=…, options_var=…, margin_type=…) - added:
function test_scale_oneway(data, method=…, center=…, transform=…, trim_frac_mean=…, trim_frac_anova=…) - added:
function wellek_to_f2(eps, n_groups)
- added:
function normal_power_het(diff, nobs, alpha, std_null=…, std_alternative=…, alternative=…) - added:
function normal_sample_size_one_tail(diff, power, alpha, std_null=…, std_alternative=…)
- added:
function confint_proportions_2indep(count1, nobs1, count2, nobs2, method=…, compare=…, alpha=…, correction=…) - added:
function power_proportions_2indep(diff, prop2, nobs1, ratio=…, alpha=…, value=…, alternative=…, return_results=…) - added:
function samplesize_proportions_2indep_onetail(diff, prop2, power, ratio=…, alpha=…, value=…, alternative=…) - added:
function score_test_proportions_2indep(count1, nobs1, count2, nobs2, value=…, compare=…, alternative=…, correction=…, return_results=…) - added:
function test_proportions_2indep(count1, nobs1, count2, nobs2, value=…, method=…, compare=…, alternative=…, correction=…, return_results=…) - added:
function tost_proportions_2indep(count1, nobs1, count2, nobs2, low, upp, method=…, compare=…, correction=…)
- added:
function etest_poisson_2indep(count1, exposure1, count2, exposure2, ratio_null=…, method=…, alternative=…, ygrid=…) - added:
function test_poisson_2indep(count1, exposure1, count2, exposure2, ratio_null=…, method=…, alternative=…, etest_kwds=…) - added:
function tost_poisson_2indep(count1, exposure1, count2, exposure2, low, upp, method=…)
- added:
class TrimmedMean(object) - added:
constructor TrimmedMean(data, fraction, is_sorted=…, axis=…) - added:
function scale_transform(data, center=…, transform=…, trim_frac=…, axis=…) - added:
function trim_mean(a, proportiontocut, axis=…) - added:
function trimboth(a, proportiontocut, axis=…) - added:
method TrimmedMean.reset_fraction(frac) - added:
method TrimmedMean.ttest_mean(value=…, transform=…, alternative=…) - added:
property TrimmedMean.data_trimmed - added:
property TrimmedMean.data_winsorized - added:
property TrimmedMean.mean_trimmed - added:
property TrimmedMean.mean_winsorized - added:
property TrimmedMean.std_mean_trimmed - added:
property TrimmedMean.std_mean_winsorized - added:
property TrimmedMean.var_winsorized
- removed:
class ParseError(Exception) - changed:
function strip_blank_lines(l) → function strip_blank_lines(line)
- added:
class InfeasibleTestError(RuntimeError) - added:
class ParseError(Exception)
- changed:
function pinv_extended(X, rcond=…) → function pinv_extended(x, rcond=…) - added:
function matrix_rank(m, tol=…, method=…)
- added:
export required_int_like
- changed:
function int_like(value, name, optional=…, strict=…) → function int_like(value: Any, name: str, optional: bool=…, strict: bool=…) -> Optional[int] - added:
function required_int_like(value: Any, name: str, strict: bool=…) -> int
- removed:
export periodogram - added:
export DynamicFactorMQ - added:
export ETSModel - added:
export STLForecast
- changed:
constructor AutoReg(endog, lags, trend=…, seasonal=…, exog=…, hold_back=…, period=…, missing=…) → constructor AutoReg(endog, lags, trend=…, seasonal=…, exog=…, hold_back=…, period=…, missing=…, *, deterministic=…, old_names=…)
- changed:
function durbin_levinson(endog, ar_order=…, demean=…, unbiased=…) → function durbin_levinson(endog, ar_order=…, demean=…, adjusted=…)
- changed:
function yule_walker(endog, ar_order=…, demean=…, unbiased=…) → function yule_walker(endog, ar_order=…, demean=…, adjusted=…)
- changed:
constructor ARIMA(endog, exog=…, order=…, seasonal_order=…, trend=…, enforce_stationarity=…, enforce_invertibility=…, concentrate_scale=…, trend_offset=…, dates=…, freq=…, missing=…) → constructor ARIMA(endog, exog=…, order=…, seasonal_order=…, trend=…, enforce_stationarity=…, enforce_invertibility=…, concentrate_scale=…, trend_offset=…, dates=…, freq=…, missing=…, validate_specification=…) - added:
method ARIMAResults.append(endog, exog=…, refit=…, fit_kwargs=…, **kwargs)
- changed:
constructor SARIMAXSpecification(endog=…, exog=…, order=…, seasonal_order=…, ar_order=…, diff=…, ma_order=…, seasonal_ar_order=…, seasonal_diff=…, seasonal_ma_order=…, seasonal_periods=…, trend=…, enforce_stationarity=…, enforce_invertibility=…, concentrate_scale=…, trend_offset=…, dates=…, freq=…, missing=…) → constructor SARIMAXSpecification(endog=…, exog=…, order=…, seasonal_order=…, ar_order=…, diff=…, ma_order=…, seasonal_ar_order=…, seasonal_diff=…, seasonal_ma_order=…, seasonal_periods=…, trend=…, enforce_stationarity=…, enforce_invertibility=…, concentrate_scale=…, trend_offset=…, dates=…, freq=…, missing=…, validate_specification=…)
- added:
variable ARIMA_DEPRECATION_WARN
- added:
class PredictionResults(object) - added:
constructor PredictionResults(predicted_mean, var_pred_mean, dist=…, df=…, row_labels=…) - added:
method PredictionResults.conf_int(alpha=…) - added:
method PredictionResults.summary_frame(alpha=…) - added:
method PredictionResults.t_test(value=…, alternative=…) - added:
property PredictionResults.predicted_mean - added:
property PredictionResults.row_labels - added:
property PredictionResults.se_mean - added:
property PredictionResults.tvalues - added:
property PredictionResults.var_pred_mean
- added:
function get_index_label_loc(key, index, row_labels) - added:
function get_index_loc(key, index) - added:
function get_prediction_index(start, end, nobs, base_index, index=…, silent=…, index_none=…, index_generated=…, data=…)
- added:
class CalendarDeterministicTerm(DeterministicTerm, ABC) - added:
class CalendarFourier(CalendarDeterministicTerm, FourierDeterministicTerm) - added:
class CalendarSeasonality(CalendarDeterministicTerm) - added:
class CalendarTimeTrend(CalendarDeterministicTerm, TimeTrendDeterministicTerm) - added:
class DeterministicProcess - added:
class DeterministicTerm(ABC) - added:
class Fourier(FourierDeterministicTerm) - added:
class FourierDeterministicTerm(DeterministicTerm, ABC) - added:
class Seasonality(DeterministicTerm) - added:
class TimeTrend(TimeTrendDeterministicTerm) - added:
class TimeTrendDeterministicTerm(DeterministicTerm, ABC) - added:
constructor CalendarDeterministicTerm(freq: str) -> None - added:
constructor CalendarFourier(freq: str, order: int) -> None - added:
constructor CalendarSeasonality(freq: str, period: str) -> None - added:
constructor CalendarTimeTrend(freq: str, constant: bool=…, order: int=…, *, base_period: Optional[Union[str, DateLike]]=…) -> None - added:
constructor DeterministicProcess(index: Union[Sequence[Hashable], pd.Index], *, period: Optional[Union[float, int]]=…, constant: bool=…, order: int=…, seasonal: bool=…, fourier: int=…, additional_terms: Sequence[DeterministicTerm]=…, drop: bool=…) - added:
constructor Fourier(period: float, order: int) - added:
constructor FourierDeterministicTerm(order: int) -> None - added:
constructor Seasonality(period: int, initial_period: int=…) -> None - added:
constructor TimeTrend(constant: bool=…, order: int=…) -> None - added:
constructor TimeTrendDeterministicTerm(constant: bool=…, order: int=…) -> None - added:
method CalendarFourier.in_sample(index: Union[Sequence[Hashable], pd.Index]) -> pd.DataFrame - added:
method CalendarFourier.out_of_sample(steps: int, index: Union[Sequence[Hashable], pd.Index], forecast_index: Optional[Sequence[Hashable]]=…) -> pd.DataFrame - added:
method CalendarSeasonality.in_sample(index: Union[Sequence[Hashable], pd.Index]) -> pd.DataFrame - added:
method CalendarSeasonality.out_of_sample(steps: int, index: Union[Sequence[Hashable], pd.Index], forecast_index: Optional[Sequence[Hashable]]=…) -> pd.DataFrame - added:
method CalendarTimeTrend.in_sample(index: Union[Sequence[Hashable], pd.Index]) -> pd.DataFrame - added:
method CalendarTimeTrend.out_of_sample(steps: int, index: Union[Sequence[Hashable], pd.Index], forecast_index: Optional[Sequence[Hashable]]=…) -> pd.DataFrame - added:
method DeterministicProcess.in_sample() -> pd.DataFrame - added:
method DeterministicProcess.out_of_sample(steps: int, forecast_index: Optional[Union[Sequence[Hashable], pd.Index]]=…) -> pd.DataFrame - added:
method DeterministicProcess.range(start: Union[IntLike, DateLike, str], stop: Union[IntLike, DateLike, str]) -> pd.DataFrame - added:
method DeterministicTerm.in_sample(index: Sequence[Hashable]) -> pd.DataFrame - added:
method DeterministicTerm.out_of_sample(steps: int, index: Sequence[Hashable], forecast_index: Optional[Sequence[Hashable]]=…) -> pd.DataFrame - added:
method Fourier.in_sample(index: Union[Sequence[Hashable], pd.Index]) -> pd.DataFrame - added:
method Fourier.out_of_sample(steps: int, index: Union[Sequence[Hashable], pd.Index], forecast_index: Optional[Sequence[Hashable]]=…) -> pd.DataFrame - added:
method Seasonality.in_sample(index: Union[Sequence[Hashable], pd.Index]) -> pd.DataFrame - added:
method Seasonality.out_of_sample(steps: int, index: Union[Sequence[Hashable], pd.Index], forecast_index: Optional[Sequence[Hashable]]=…) -> pd.DataFrame - added:
method TimeTrend.in_sample(index: Union[Sequence[Hashable], pd.Index]) -> pd.DataFrame - added:
method TimeTrend.out_of_sample(steps: int, index: Union[Sequence[Hashable], pd.Index], forecast_index: Optional[Sequence[Hashable]]=…) -> pd.DataFrame - added:
property CalendarDeterministicTerm.freq -> str - added:
property CalendarSeasonality.freq -> str - added:
property CalendarSeasonality.period -> str - added:
property CalendarTimeTrend.base_period -> Optional[str] - added:
property DeterministicProcess.index -> pd.Index - added:
property DeterministicProcess.terms -> List[DeterministicTerm] - added:
property DeterministicTerm.is_dummy -> bool - added:
property Fourier.period -> float - added:
property FourierDeterministicTerm.order -> int - added:
property Seasonality.initial_period -> int - added:
property Seasonality.period -> int - added:
property TimeTrendDeterministicTerm.constant -> bool - added:
property TimeTrendDeterministicTerm.order -> int - added:
static method CalendarTimeTrend.from_string(freq: str, trend: str, base_period: Optional[Union[str, DateLike]]=…) -> CalendarTimeTrend - added:
static method Seasonality.from_index(index: Union[Sequence[Hashable], pd.DatetimeIndex, pd.PeriodIndex]) -> Seasonality - added:
static method TimeTrend.from_string(trend: str) -> TimeTrend - added:
variable DateLike - added:
variable IntLike - added:
variable START_BEFORE_INDEX_ERR
- added:
class StateSpaceMLEModel(tsbase.TimeSeriesModel) - added:
class StateSpaceMLEResults(tsbase.TimeSeriesModelResults) - added:
constructor StateSpaceMLEModel(endog, exog=…, dates=…, freq=…, missing=…, **kwargs) - added:
constructor StateSpaceMLEResults(model, params, scale=…) - added:
method StateSpaceMLEModel.clone(endog, exog=…, **kwargs) - added:
method StateSpaceMLEModel.fit_constrained(constraints, start_params=…, **fit_kwds) - added:
method StateSpaceMLEModel.fix_params(params) - added:
method StateSpaceMLEResults.aic() - added:
method StateSpaceMLEResults.aicc() - added:
method StateSpaceMLEResults.bic() - added:
method StateSpaceMLEResults.cov_params_approx() - added:
method StateSpaceMLEResults.df_resid() - added:
method StateSpaceMLEResults.fittedvalues() - added:
method StateSpaceMLEResults.hqic() - added:
method StateSpaceMLEResults.llf() - added:
method StateSpaceMLEResults.mae() - added:
method StateSpaceMLEResults.mse() - added:
method StateSpaceMLEResults.nobs_effective() - added:
method StateSpaceMLEResults.pvalues() - added:
method StateSpaceMLEResults.resid() - added:
method StateSpaceMLEResults.sse() - added:
method StateSpaceMLEResults.summary(alpha=…, start=…, title=…, model_name=…, display_params=…) - added:
method StateSpaceMLEResults.test_heteroskedasticity(method, alternative=…, use_f=…) - added:
method StateSpaceMLEResults.test_normality(method) - added:
method StateSpaceMLEResults.test_serial_correlation(method, lags=…) - added:
method StateSpaceMLEResults.zvalues() - added:
property StateSpaceMLEModel.k_params - added:
property StateSpaceMLEModel.param_names - added:
property StateSpaceMLEModel.start_params - added:
static method StateSpaceMLEModel.from_formula(formula, data, subset=…, drop_cols=…, *args, **kwargs) - added:
static method StateSpaceMLEModel.prepare_data(data)
- added:
class ETSModel(base.StateSpaceMLEModel) - added:
class ETSResults(base.StateSpaceMLEResults) - added:
class ETSResultsWrapper(wrap.ResultsWrapper) - added:
class PredictionResults - added:
class PredictionResultsWrapper(wrap.ResultsWrapper) - added:
constructor ETSModel(endog, error=…, trend=…, damped_trend=…, seasonal=…, seasonal_periods=…, initialization_method=…, initial_level=…, initial_trend=…, initial_seasonal=…, bounds=…, dates=…, freq=…, missing=…) - added:
constructor ETSResults(model, params, results) - added:
constructor PredictionResults(results, start=…, end=…, dynamic=…, index=…, method=…, simulate_repetitions=…, **simulate_kwargs) - added:
method ETSModel.fit(start_params=…, maxiter=…, full_output=…, disp=…, callback=…, return_params=…, **kwargs) - added:
method ETSModel.hessian(params, approx_centered=…, approx_complex_step=…, **kwargs) - added:
method ETSModel.loglike(params, **kwargs) - added:
method ETSModel.score(params, approx_centered=…, approx_complex_step=…, **kwargs) - added:
method ETSModel.set_bounds(bounds) - added:
method ETSModel.set_initialization_method(initialization_method, initial_level=…, initial_trend=…, initial_seasonal=…) - added:
method ETSModel.smooth(params, return_raw=…) - added:
method ETSModel.update(*args, **kwargs) - added:
method ETSModel.use_internal_loglike() - added:
method ETSResults.fittedvalues() - added:
method ETSResults.forecast(steps=…) - added:
method ETSResults.get_prediction(start=…, end=…, dynamic=…, index=…, method=…, simulate_repetitions=…, **simulate_kwargs) - added:
method ETSResults.llf() - added:
method ETSResults.nobs_effective() - added:
method ETSResults.predict(start=…, end=…, dynamic=…, index=…) - added:
method ETSResults.resid() - added:
method ETSResults.simulate(nsimulations, anchor=…, repetitions=…, random_errors=…, random_state=…) - added:
method ETSResults.summary(alpha=…, start=…) - added:
method PredictionResults.pred_int(alpha=…) - added:
method PredictionResults.summary_frame(endog=…, alpha=…) - added:
property ETSModel.initial_state_names - added:
property ETSModel.k_endog - added:
property ETSModel.k_params - added:
property ETSModel.nobs_effective - added:
property ETSModel.short_name - added:
property ETSModel.state_names - added:
property PredictionResults.var_pred_mean - added:
static method ETSModel.prepare_data(data)
- added:
class STLForecast - added:
class STLForecastResults - added:
constructor STLForecast(endog, model, *, model_kwargs=…, period=…, seasonal=…, trend=…, low_pass=…, seasonal_deg=…, trend_deg=…, low_pass_deg=…, robust=…, seasonal_jump=…, trend_jump=…, low_pass_jump=…) - added:
constructor STLForecastResults(stl: STL, result: DecomposeResult, model, model_result, endog) -> None - added:
method STLForecast.fit(*, inner_iter=…, outer_iter=…, fit_kwargs=…) - added:
method STLForecastResults.forecast(steps: int=…, **kwargs: Dict[str, Any]) -> Union[np.ndarray, pd.Series] - added:
method STLForecastResults.get_prediction(start: Optional[DateLike]=…, end: Optional[DateLike]=…, dynamic: Union[bool, DateLike]=…, **kwargs: Dict[str, Any]) - added:
method STLForecastResults.summary() -> Summary - added:
property STLForecastResults.model -> Any - added:
property STLForecastResults.model_result -> Any - added:
property STLForecastResults.period -> int - added:
property STLForecastResults.result -> DecomposeResult - added:
property STLForecastResults.stl -> STL - added:
variable DateLike - added:
variable ds
- added:
class ThetaModel - added:
class ThetaModelResults - added:
constructor ThetaModel(endog, *, period: Optional[int]=…, deseasonalize: bool=…, use_test: bool=…, method: str=…, difference: bool=…) -> None - added:
constructor ThetaModelResults(b0: float, alpha: float, sigma2: Optional[float], one_step: float, seasonal: np.ndarray, use_mle: bool, model: ThetaModel) -> None - added:
function extend_index(steps: int, index: pd.Index) -> pd.Index - added:
method ThetaModel.fit(use_mle: bool=…, disp: bool=…) -> ThetaModelResults - added:
method ThetaModelResults.forecast(steps: int=…, theta: float=…) -> pd.Series - added:
method ThetaModelResults.forecast_components(steps: int=…) -> pd.DataFrame - added:
method ThetaModelResults.plot_predict(steps: int=…, theta: float=…, alpha: Optional[float]=…, in_sample: bool=…, fig: Optional[matplotlib.figure.Figure]=…, figsize: Tuple[float, float]=…) -> matplotlib.figure.Figure - added:
method ThetaModelResults.prediction_intervals(steps: int=…, theta: float=…, alpha: float=…) -> pd.DataFrame - added:
method ThetaModelResults.summary() -> Summary - added:
property ThetaModel.deseasonalize -> bool - added:
property ThetaModel.difference -> bool - added:
property ThetaModel.method -> str - added:
property ThetaModel.period -> int - added:
property ThetaModel.use_test -> bool - added:
property ThetaModelResults.model -> ThetaModel - added:
property ThetaModelResults.params -> pd.Series - added:
property ThetaModelResults.sigma2 -> float
- removed:
class ExponentialSmoothing(TimeSeriesModel) - removed:
class Holt(ExponentialSmoothing) - removed:
class HoltWintersResults(Results) - removed:
class HoltWintersResultsWrapper(ResultsWrapper) - removed:
class SimpleExpSmoothing(ExponentialSmoothing) - removed:
constructor ExponentialSmoothing(endog, trend=…, damped=…, seasonal=…, seasonal_periods=…, dates=…, freq=…, missing=…) - removed:
constructor Holt(endog, exponential=…, damped=…) - removed:
constructor HoltWintersResults(model, params, **kwargs) - removed:
constructor SimpleExpSmoothing(endog) - removed:
method ExponentialSmoothing.fit(smoothing_level=…, smoothing_slope=…, smoothing_seasonal=…, damping_slope=…, optimized=…, use_boxcox=…, remove_bias=…, use_basinhopping=…, start_params=…, initial_level=…, initial_slope=…, use_brute=…) - removed:
method ExponentialSmoothing.initial_values() - removed:
method ExponentialSmoothing.predict(params, start=…, end=…) - removed:
method Holt.fit(smoothing_level=…, smoothing_slope=…, damping_slope=…, optimized=…, start_params=…, initial_level=…, initial_slope=…, use_brute=…) - removed:
method HoltWintersResults.forecast(steps=…) - removed:
method HoltWintersResults.predict(start=…, end=…) - removed:
method HoltWintersResults.summary() - removed:
method SimpleExpSmoothing.fit(smoothing_level=…, optimized=…, start_params=…, initial_level=…, use_brute=…) - removed:
variable PY_SMOOTHERS - removed:
variable SMOOTHERS - added:
export ExponentialSmoothing - added:
export Holt - added:
export HoltWintersResults - added:
export PY_SMOOTHERS - added:
export SMOOTHERS - added:
export SimpleExpSmoothing
- added:
class ExponentialSmoothing(TimeSeriesModel) - added:
class Holt(ExponentialSmoothing) - added:
class SimpleExpSmoothing(ExponentialSmoothing) - added:
constructor ExponentialSmoothing(endog, trend=…, damped_trend=…, seasonal=…, *, seasonal_periods=…, initialization_method=…, initial_level=…, initial_trend=…, initial_seasonal=…, use_boxcox=…, bounds=…, dates=…, freq=…, missing=…) - added:
constructor Holt(endog, exponential=…, damped_trend=…, initialization_method=…, initial_level=…, initial_trend=…) - added:
constructor SimpleExpSmoothing(endog, initialization_method=…, initial_level=…) - added:
function opt_wrapper(func) - added:
method ExponentialSmoothing.fit(smoothing_level=…, smoothing_trend=…, smoothing_seasonal=…, damping_trend=…, *, optimized=…, remove_bias=…, start_params=…, method=…, minimize_kwargs=…, use_brute=…, use_boxcox=…, use_basinhopping=…, initial_level=…, initial_trend=…) - added:
method ExponentialSmoothing.fix_params(values) - added:
method ExponentialSmoothing.initial_values(initial_level=…, initial_trend=…, force=…) - added:
method ExponentialSmoothing.predict(params, start=…, end=…) - added:
method Holt.fit(smoothing_level=…, smoothing_trend=…, *, damping_trend=…, optimized=…, start_params=…, initial_level=…, initial_trend=…, use_brute=…, use_boxcox=…, remove_bias=…, method=…, minimize_kwargs=…) - added:
method SimpleExpSmoothing.fit(smoothing_level=…, *, optimized=…, start_params=…, initial_level=…, use_brute=…, use_boxcox=…, remove_bias=…, method=…, minimize_kwargs=…) - added:
variable PY_SMOOTHERS - added:
variable SMOOTHERS
- added:
class HoltWintersResults(Results) - added:
class HoltWintersResultsWrapper(ResultsWrapper) - added:
constructor HoltWintersResults(model, params, sse, aic, aicc, bic, optimized, level, trend, season, params_formatted, resid, k, fittedvalues, fittedfcast, fcastvalues, mle_retvals=…) - added:
method HoltWintersResults.forecast(steps=…) - added:
method HoltWintersResults.predict(start=…, end=…) - added:
method HoltWintersResults.simulate(nsimulations, anchor=…, repetitions=…, error=…, random_errors=…, random_state=…) - added:
method HoltWintersResults.summary() - added:
property HoltWintersResults.aic - added:
property HoltWintersResults.aicc - added:
property HoltWintersResults.bic - added:
property HoltWintersResults.fcastvalues - added:
property HoltWintersResults.fittedfcast - added:
property HoltWintersResults.fittedvalues - added:
property HoltWintersResults.k - added:
property HoltWintersResults.level - added:
property HoltWintersResults.mle_retvals - added:
property HoltWintersResults.model - added:
property HoltWintersResults.optimized - added:
property HoltWintersResults.params_formatted - added:
property HoltWintersResults.resid - added:
property HoltWintersResults.season - added:
property HoltWintersResults.slope - added:
property HoltWintersResults.sse - added:
property HoltWintersResults.trend
- added:
variable NON_STATIONARY_ERROR
- added:
class CFASimulationSmoother(object) - added:
constructor CFASimulationSmoother(model, cfa_simulation_smoother_classes=…) - added:
method CFASimulationSmoother.simulate(variates=…, update_posterior=…) - added:
property CFASimulationSmoother.posterior_cov - added:
property CFASimulationSmoother.posterior_cov_inv_chol_sparse - added:
property CFASimulationSmoother.posterior_mean
- added:
class DynamicFactorMQ(mlemodel.MLEModel) - added:
class DynamicFactorMQResults(mlemodel.MLEResults) - added:
class DynamicFactorMQStates(dict) - added:
class FactorBlock(dict) - added:
constructor DynamicFactorMQ(endog, k_endog_monthly=…, factors=…, factor_orders=…, factor_multiplicities=…, idiosyncratic_ar1=…, standardize=…, endog_quarterly=…, init_t0=…, obs_cov_diag=…, **kwargs) - added:
constructor DynamicFactorMQResults(model, params, filter_results, cov_type=…, **kwargs) - added:
constructor DynamicFactorMQStates(k_endog_M, k_endog_Q, endog_names, factors, factor_orders, factor_multiplicities, idiosyncratic_ar1) - added:
constructor FactorBlock(factor_names, factor_order, endog_factor_map, state_offset, k_endog_Q) - added:
method DynamicFactorMQ.clone(endog, k_endog_monthly=…, endog_quarterly=…, retain_standardization=…, **kwargs) - added:
method DynamicFactorMQ.filter(params, transformed=…, includes_fixed=…, complex_step=…, cov_type=…, cov_kwds=…, return_ssm=…, results_class=…, results_wrapper_class=…, low_memory=…, **kwargs) - added:
method DynamicFactorMQ.fit(start_params=…, transformed=…, includes_fixed=…, cov_type=…, cov_kwds=…, method=…, maxiter=…, tolerance=…, em_initialization=…, mstep_method=…, full_output=…, disp=…, callback=…, return_params=…, optim_score=…, optim_complex_step=…, optim_hessian=…, flags=…, low_memory=…, llf_decrease_action=…, llf_decrease_tolerance=…, **kwargs) - added:
method DynamicFactorMQ.fit_em(start_params=…, transformed=…, cov_type=…, cov_kwds=…, maxiter=…, tolerance=…, disp=…, em_initialization=…, mstep_method=…, full_output=…, return_params=…, low_memory=…, llf_decrease_action=…, llf_decrease_tolerance=…) - added:
method DynamicFactorMQ.impulse_responses(params, steps=…, impulse=…, orthogonalized=…, cumulative=…, anchor=…, exog=…, extend_model=…, extend_kwargs=…, transformed=…, includes_fixed=…, original_scale=…, **kwargs) - added:
method DynamicFactorMQ.loading_constraints(i) - added:
method DynamicFactorMQ.simulate(params, nsimulations, measurement_shocks=…, state_shocks=…, initial_state=…, anchor=…, repetitions=…, exog=…, extend_model=…, extend_kwargs=…, transformed=…, includes_fixed=…, original_scale=…, **kwargs) - added:
method DynamicFactorMQ.smooth(params, transformed=…, includes_fixed=…, complex_step=…, cov_type=…, cov_kwds=…, return_ssm=…, results_class=…, results_wrapper_class=…, **kwargs) - added:
method DynamicFactorMQ.summary(truncate_endog_names=…) - added:
method DynamicFactorMQ.transform_params(unconstrained) - added:
method DynamicFactorMQ.untransform_params(constrained) - added:
method DynamicFactorMQ.update(params, **kwargs) - added:
method DynamicFactorMQResults.append(endog, endog_quarterly=…, refit=…, fit_kwargs=…, copy_initialization=…, retain_standardization=…, **kwargs) - added:
method DynamicFactorMQResults.apply(endog, k_endog_monthly=…, endog_quarterly=…, refit=…, fit_kwargs=…, copy_initialization=…, retain_standardization=…, **kwargs) - added:
method DynamicFactorMQResults.coefficients_of_determination() - added:
method DynamicFactorMQResults.extend(endog, endog_quarterly=…, fit_kwargs=…, retain_standardization=…, **kwargs) - added:
method DynamicFactorMQResults.get_coefficients_of_determination(method=…, which=…) - added:
method DynamicFactorMQResults.get_prediction(start=…, end=…, dynamic=…, index=…, exog=…, extend_model=…, extend_kwargs=…, original_scale=…, **kwargs) - added:
method DynamicFactorMQResults.news(comparison, impact_date=…, impacted_variable=…, start=…, end=…, periods=…, exog=…, comparison_type=…, return_raw=…, tolerance=…, endog_quarterly=…, original_scale=…, **kwargs) - added:
method DynamicFactorMQResults.plot_coefficients_of_determination(method=…, which=…, endog_labels=…, fig=…, figsize=…) - added:
method DynamicFactorMQResults.summary(alpha=…, start=…, title=…, model_name=…, display_params=…, display_diagnostics=…, display_params_as_list=…, truncate_endog_names=…, display_max_endog=…) - added:
property DynamicFactorMQ.loglike_constant - added:
property DynamicFactorMQ.param_names - added:
property DynamicFactorMQ.start_params - added:
property DynamicFactorMQ.state_names - added:
property DynamicFactorMQResults.factors - added:
property DynamicFactorMQStates.endog_factor_iloc - added:
property DynamicFactorMQStates.factors_L1 - added:
property DynamicFactorMQStates.factors_L1_5_ix - added:
property DynamicFactorMQStates.idio_ar_L1 - added:
property DynamicFactorMQStates.idio_ar_M - added:
property DynamicFactorMQStates.idio_ar_Q - added:
property DynamicFactorMQStates.idio_ar_Q_ix - added:
property FactorBlock.factors - added:
property FactorBlock.factors_L1 - added:
property FactorBlock.factors_L1_5 - added:
property FactorBlock.factors_ar - added:
property FactorBlock.factors_ix - added:
static method DynamicFactorMQ.construct_endog(endog_monthly, endog_quarterly)
- added:
field KalmanFilter.filter_chandrasekhar - added:
variable FILTER_CHANDRASEKHAR
- changed:
method KalmanSmoother.smooth(smoother_output=…, smooth_method=…, results=…, run_filter=…, prefix=…, complex_step=…, **kwargs) → method KalmanSmoother.smooth(smoother_output=…, smooth_method=…, results=…, run_filter=…, prefix=…, complex_step=…, update_representation=…, update_filter=…, update_smoother=…, **kwargs) - added:
method SmootherResults.news(previous, t=…, start=…, end=…, revised=…, design=…) - added:
method SmootherResults.smoothed_state_autocovariance(lag=…, t=…, start=…, end=…, extend_kwargs=…) - added:
method SmootherResults.smoothed_state_gain(updates_ix, t=…, start=…, end=…, extend_kwargs=…)
- changed:
method MLEResults.append(endog, exog=…, refit=…, fit_kwargs=…, **kwargs) → method MLEResults.append(endog, exog=…, refit=…, fit_kwargs=…, copy_initialization=…, **kwargs) - changed:
method MLEResults.apply(endog, exog=…, refit=…, fit_kwargs=…, **kwargs) → method MLEResults.apply(endog, exog=…, refit=…, fit_kwargs=…, copy_initialization=…, **kwargs) - changed:
method MLEResults.plot_diagnostics(variable=…, lags=…, fig=…, figsize=…) → method MLEResults.plot_diagnostics(variable=…, lags=…, fig=…, figsize=…, truncate_endog_names=…) - changed:
method MLEResults.summary(alpha=…, start=…, title=…, model_name=…, display_params=…) → method MLEResults.summary(alpha=…, start=…, title=…, model_name=…, display_params=…, display_diagnostics=…, truncate_endog_names=…, display_max_endog=…, extra_top_left=…, extra_top_right=…) - changed:
method PredictionResults.summary_frame(endog=…, what=…, alpha=…) → method PredictionResults.summary_frame(endog=…, alpha=…) - added:
method MLEResults.news(comparison, impact_date=…, impacted_variable=…, start=…, end=…, periods=…, exog=…, comparison_type=…, return_raw=…, tolerance=…, **kwargs)
- added:
class NewsResults(object) - added:
constructor NewsResults(news_results, model, updated, previous, impacted_variable=…, tolerance=…, row_labels=…) - added:
method NewsResults.summary(impact_date=…, impacted_variable=…, update_date=…, updated_variable=…, impacts_groupby=…, details_groupby=…, show_revisions_columns=…, sparsify=…, include_details_tables=…, include_revisions_tables=…, float_format=…) - added:
method NewsResults.summary_details(impact_date=…, impacted_variable=…, update_date=…, updated_variable=…, groupby=…, sparsify=…, float_format=…, multiple_tables=…) - added:
method NewsResults.summary_impacts(impact_date=…, impacted_variable=…, groupby=…, show_revisions_columns=…, sparsify=…, float_format=…) - added:
method NewsResults.summary_news(sparsify=…) - added:
method NewsResults.summary_revisions(sparsify=…) - added:
property NewsResults.data_revisions - added:
property NewsResults.data_updates - added:
property NewsResults.details_by_impact - added:
property NewsResults.details_by_update - added:
property NewsResults.impacted_variable - added:
property NewsResults.impacts - added:
property NewsResults.tolerance
- added:
method Representation.diff_endog(new_endog, tolerance=…)
- changed:
constructor SARIMAX(endog, exog=…, order=…, seasonal_order=…, trend=…, measurement_error=…, time_varying_regression=…, mle_regression=…, simple_differencing=…, enforce_stationarity=…, enforce_invertibility=…, hamilton_representation=…, concentrate_scale=…, trend_offset=…, use_exact_diffuse=…, dates=…, freq=…, missing=…, **kwargs) → constructor SARIMAX(endog, exog=…, order=…, seasonal_order=…, trend=…, measurement_error=…, time_varying_regression=…, mle_regression=…, simple_differencing=…, enforce_stationarity=…, enforce_invertibility=…, hamilton_representation=…, concentrate_scale=…, trend_offset=…, use_exact_diffuse=…, dates=…, freq=…, missing=…, validate_specification=…, **kwargs)
- changed:
method SimulationSmoother.simulation_smoother(simulation_output=…, results_class=…, prefix=…, **kwargs) → method SimulationSmoother.simulation_smoother(simulation_output=…, method=…, results_class=…, prefix=…, **kwargs)
- added:
variable prefix_cfa_simulation_smoother_map
- removed:
function periodogram(x) - changed:
function acf(x, unbiased=…, nlags=…, qstat=…, fft=…, alpha=…, missing=…) → function acf(x, adjusted=…, nlags=…, qstat=…, fft=…, alpha=…, missing=…) - changed:
function acovf(x, unbiased=…, demean=…, fft=…, missing=…, nlag=…) → function acovf(x, adjusted=…, demean=…, fft=…, missing=…, nlag=…) - changed:
function ccf(x, y, unbiased=…) → function ccf(x, y, adjusted=…) - changed:
function ccovf(x, y, unbiased=…, demean=…) → function ccovf(x, y, adjusted=…, demean=…) - changed:
function pacf_ols(x, nlags=…, efficient=…, unbiased=…) → function pacf_ols(x, nlags=…, efficient=…, adjusted=…)
- changed:
method BaseIRAnalysis.plot(orth=…, impulse=…, response=…, signif=…, plot_params=…, subplot_params=…, plot_stderr=…, stderr_type=…, repl=…, seed=…, component=…) → method BaseIRAnalysis.plot(orth=…, *, impulse=…, response=…, signif=…, plot_params=…, figsize=…, subplot_params=…, plot_stderr=…, stderr_type=…, repl=…, seed=…, component=…) - changed:
method BaseIRAnalysis.plot_cum_effects(orth=…, impulse=…, response=…, signif=…, plot_params=…, subplot_params=…, plot_stderr=…, stderr_type=…, repl=…, seed=…) → method BaseIRAnalysis.plot_cum_effects(orth=…, *, impulse=…, response=…, signif=…, plot_params=…, figsize=…, subplot_params=…, plot_stderr=…, stderr_type=…, repl=…, seed=…)
- added:
static method VAR.from_formula(formula, data, subset=…, drop_cols=…, *args, **kwargs)