naginterfaces.library.tsa.uni_garch_asym1_estim¶
- naginterfaces.library.tsa.uni_garch_asym1_estim(dist, yt, x, ip, iq, mn, isym, theta, pht, copts, maxit, tol, io_manager=None)[source]¶
uni_garch_asym1_estim
estimates the parameters of either a standard univariate regression GARCH process, or a univariate regression-type I process (see Engle and Ng (1993)).For full information please refer to the NAG Library document for g13fa
https://support.nag.com/numeric/nl/nagdoc_30.2/flhtml/g13/g13faf.html
- Parameters
- diststr, length 1
The type of distribution to use for .
A Normal distribution is used.
A Student’s -distribution is used.
- ytfloat, array-like, shape
The sequence of observations, , for .
- xfloat, array-like, shape
Row of must contain the time dependent exogenous vector , where , for .
- ipint
The number of coefficients, , for .
- iqint
The number of coefficients, , for .
- mnint
If , the mean term will be included in the model.
- isymint
If , the asymmetry term will be included in the model.
- thetafloat, array-like, shape
The initial parameter estimates for the vector .
The first element must contain the coefficient and the next elements must contain the coefficients , for .
The next elements must contain the coefficients , for .
If , the next element must contain the asymmetry parameter .
If , the next element must contain , the number of degrees of freedom of the Student’s -distribution.
If , the next element must contain the mean term .
If , the remaining elements are taken as initial estimates of the linear regression coefficients , for .
- phtfloat
If , is the value to be used for the pre-observed conditional variance; otherwise is not referenced.
- coptsbool, array-like, shape
The options to be used by
uni_garch_asym1_estim
.Stationary conditions are enforced, otherwise they are not.
The function provides initial parameter estimates of the regression terms, otherwise these are to be provided by you.
- maxitint
The maximum number of iterations to be used by the optimization function when estimating the parameters. If is set to , the standard errors, score vector and variance-covariance are calculated for the input value of in when ; however the value of is not updated.
- tolfloat
The tolerance to be used by the optimization function when estimating the parameters.
- io_managerFileObjManager, optional
Manager for I/O in this routine.
- Returns
- thetafloat, ndarray, shape
The estimated values for the vector .
The first element contains the coefficient , the next elements contain the coefficients , for .
The next elements are the coefficients , for .
If , the next element contains the estimate for the asymmetry parameter .
If , the next element contains an estimate for , the number of degrees of freedom of the Student’s -distribution.
If , the next element contains an estimate for the mean term .
The final elements are the estimated linear regression coefficients , for .
- sefloat, ndarray, shape
The standard errors for .
The first element contains the standard error for .
The next elements contain the standard errors for , for .
The next elements are the standard errors for , for .
If , the next element contains the standard error for .
If , the next element contains the standard error for , the number of degrees of freedom of the Student’s -distribution.
If , the next element contains the standard error for .
The final elements are the standard errors for , for .
- scfloat, ndarray, shape
The scores for .
The first element contains the score for .
The next elements contain the score for , for .
The next elements are the scores for , for .
If , the next element contains the score for .
If , the next element contains the score for , the number of degrees of freedom of the Student’s -distribution.
If , the next element contains the score for .
The final elements are the scores for , for .
- covrfloat, ndarray, shape
The covariance matrix of the parameter estimates , that is the inverse of the Fisher Information Matrix.
- phtfloat
If , is the estimated value of the pre-observed conditional variance.
- etfloat, ndarray, shape
The estimated residuals, , for .
- htfloat, ndarray, shape
The estimated conditional variances, , for .
- lgffloat
The value of the log-likelihood function at .
- Raises
- NagValueError
- (errno )
On entry, .
Constraint: .
- (errno )
On entry, .
Constraint: .
- (errno )
On entry, .
Constraint: if then , else .
- (errno )
On entry, .
Constraint: .
- (errno )
On entry, .
Constraint: .
- (errno )
On entry, .
Constraint: or .
- (errno )
On entry, .
Constraint: .
- (errno )
On entry, .
Constraint: .
- (errno )
On entry, .
Constraint: or .
- (errno )
On entry, .
Constraint: or .
- (errno )
On entry, .
Constraint: .
- (errno )
On entry, the matrix is not full rank.
- (errno )
The information matrix is not positive definite.
- (errno )
The maximum number of iterations has been reached.
- (errno )
No feasible model parameters could be found.
- Warns
- NagAlgorithmicWarning
- (errno )
The log-likelihood cannot be optimized any further.
- Notes
In the NAG Library the traditional C interface for this routine uses a different algorithmic base. Please contact NAG if you have any questions about compatibility.
A univariate regression-type I process, with coefficients , for , coefficients , for , and linear regression coefficients , for , can be represented by:
where or . Here is a standardized Student’s -distribution with degrees of freedom and variance , is the number of terms in the sequence, denotes the endogenous variables, the exogenous variables, the regression mean, the regression coefficients, the residuals, the conditional variance, the number of degrees of freedom of the Student’s -distribution, and the set of all information up to time .
uni_garch_asym1_estim
provides an estimate for , the parameter vector where , when and when ., and can be used to simplify the expression in (1) as follows:
No Regression and No Mean
,
,
,
and
is a vector when and a vector when .
No Regression
,
,
,
and
is a vector when and a vector when .
Note: if the , where is known (not to be estimated by
uni_garch_asym1_estim
) then (1) can be written as , where . This corresponds to the case No Regression and No Mean, with replaced by .No Mean
,
,
,
and
is a vector when and a vector when .
- References
Bollerslev, T, 1986, Generalised autoregressive conditional heteroskedasticity, Journal of Econometrics (31), 307–327
Engle, R, 1982, Autoregressive conditional heteroskedasticity with estimates of the variance of United Kingdom inflation, Econometrica (50), 987–1008
Engle, R and Ng, V, 1993, Measuring and testing the impact of news on volatility, Journal of Finance (48), 1749–1777
Hamilton, J, 1994, Time Series Analysis, Princeton University Press