arch.univariate.base.ARCHModelResult¶

class arch.univariate.base.ARCHModelResult(params: ndarray[tuple[int], dtype[float64]], param_cov: ndarray[tuple[int, ...], dtype[float64]] | None, r2: float, resid: ndarray[tuple[int], dtype[float64]], volatility: ndarray[tuple[int], dtype[float64]], cov_type: str, dep_var: Series, names: list[str], loglikelihood: float, is_pandas: bool, optim_output: OptimizeResult, fit_start: int, fit_stop: int, model: ARCHModel)[source]¶

Results from estimation of an ARCHModel model.

Parameters:¶

params: ndarray[tuple[int], dtype[float64]]¶

Estimated parameters.

param_cov: ndarray[tuple[int, ...], dtype[float64]] | None¶

Estimated variance-covariance matrix of params. If none, calls method to compute variance from model when parameter covariance is first used from result.

r2: float¶

Model R-squared.

resid: ndarray[tuple[int], dtype[float64]]¶

Residuals from model. Residuals have same shape as original data and contain nan-values in locations not used in estimation.

volatility: ndarray[tuple[int], dtype[float64]]¶

Conditional volatility from model.

cov_type: str¶

String describing the covariance estimator used.

dep_var: Series¶

Dependent variable.

names: list[str]¶

Model parameter names.

loglikelihood: float¶

Loglikelihood at estimated parameters.

is_pandas: bool¶

Whether the original input was pandas.

optim_output: OptimizeResult¶

Result of log-likelihood optimization.

fit_start: int¶

Integer index of the first observation used to fit the model.

fit_stop: int¶

Integer index of the last observation used to fit the model using slice notation fit_start:fit_stop.

model: ARCHModel¶

The model object used to estimate the parameters.

Notes

The results are created by ARCHModel.fit. The arguments of the constructor are not validated.

Methods

arch_lm_test([lags, standardized])	Perform an ARCH LM test for conditional heteroskedasticity.
conf_int([alpha])	Compute parameter confidence intervals.
forecast([params, horizon, start, align, ...])	Construct forecasts from estimated model.
hedgehog_plot([params, horizon, step, ...])	Plot forecasts from estimated model.
hill_estimator([k])	Compute the Hill tail-index estimator of standardized residuals.
plot([annualize, scale])	Plot standardized residuals and conditional volatility.
summary()	Construct a summary of the results from a fit model.
var_ratio([level])	Compute the ratio of Gaussian VaR to model-distribution VaR.

Properties

aic	Akaike Information Criteria.
bic	Schwarz/Bayesian Information Criteria.
conditional_volatility	Estimated conditional volatility.
convergence_flag	The scipy.optimize.minimize result flag.
excess_kurtosis	Sample excess kurtosis of standardized residuals.
fit_start	Start of sample used to estimate parameters.
fit_stop	End of sample used to estimate parameters.
loglikelihood	Model loglikelihood.
model	Model instance used to produce the fit.
nobs	Number of data points used to estimate model.
num_params	Number of parameters in model.
optimization_result	Information about the convergence of the loglikelihood optimization.
param_cov	Parameter covariance.
params	Model parameters.
pvalues	Array of p-values for the t-statistics.
resid	Model residuals.
rsquared	R-squared.
rsquared_adj	Degree of freedom adjusted R-squared.
scale	The scale applied to the original data before estimating the model.
std_err	Array of parameter standard errors.
std_resid	Standardized residuals, the residuals divided by the conditional volatility.
tvalues	Array of t-statistics testing the null that the coefficient are 0.