A multivariate time varying autoregressive modeling of nonstationary covariance time series

1983 ◽  
Author(s):  
Will Gersch ◽  
Alan Gevins ◽  
Genshiro Kitagawa
Keyword(s):  
Time Series ◽  
Time Varying ◽  
Autoregressive Modeling ◽  
Nonstationary Covariance

COMPARING TIME-VARYING AUTOREGRESSIVE STRUCTURES OF LOCALLY STATIONARY PROCESSES

2008 ◽  
Vol 06 (01) ◽  
pp. 1-23 ◽  
Author(s):  
GLADYS E. SALCEDO ◽  
JOÃO R. SATO ◽  
PEDRO A. MORETTIN ◽  
CLÉLIA M. TOLOI
Keyword(s):  
Time Series ◽  
Financial Time Series ◽  
Covariance Structure ◽  
Wald Test ◽  
Stationary Time Series ◽  
Time Varying ◽  
Simulation Studies ◽  
Autoregressive Modeling ◽  
Locally Stationary Processes ◽  
And Function

In this paper, a novel statistical test is introduced to compare two locally stationary time series. The proposed approach is a Wald test considering time-varying autoregressive modeling and function projections in adequate spaces. The covariance structure of the innovations may be also time-varying. In order to obtain function estimators for the time-varying autoregressive parameters, we consider function expansions in splines and wavelet bases. Simulation studies provide evidence that the proposed test has a good performance. We also assess its usefulness when applied to a financial time series.

Selective Linear Segmentation for Detecting Relevant Parameter Changes*

2020 ◽  
Author(s):  
Arnaud Dufays ◽  
Elysee Aristide Houndetoungan ◽  
Alain Coën
Keyword(s):  
Time Series ◽  
Hedge Fund ◽  
Expectation Maximization Algorithm ◽  
Model Parameters ◽  
Relevant Parameter ◽  
Time Varying ◽  
Long Time ◽  
Model Selection Uncertainty ◽  
Penalized Likelihood Approach ◽  
Changes Over Time

Abstract Change-point (CP) processes are one flexible approach to model long time series. We propose a method to uncover which model parameters truly vary when a CP is detected. Given a set of breakpoints, we use a penalized likelihood approach to select the best set of parameters that changes over time and we prove that the penalty function leads to a consistent selection of the true model. Estimation is carried out via the deterministic annealing expectation-maximization algorithm. Our method accounts for model selection uncertainty and associates a probability to all the possible time-varying parameter specifications. Monte Carlo simulations highlight that the method works well for many time series models including heteroskedastic processes. For a sample of fourteen hedge fund (HF) strategies, using an asset-based style pricing model, we shed light on the promising ability of our method to detect the time-varying dynamics of risk exposures as well as to forecast HF returns.

scholarly journals An Extended Time Series Algorithm for Modal Identification from Nonstationary Ambient Response Data Only

2014 ◽  
Vol 2014 ◽  
pp. 1-12 ◽  
Author(s):  
Chang-Sheng Lin ◽  
Dar-Yun Chiang ◽  
Tse-Chuan Tseng
Keyword(s):  
Time Series ◽  
Arma Model ◽  
Identification Problem ◽  
Nonstationary Time Series ◽  
Modal Identification ◽  
Ambient Vibration ◽  
Time Varying ◽  
Varying Parameters ◽  
Response Data ◽  
Time Varying Parameters

Modal Identification is considered from response data of structural systems under nonstationary ambient vibration. The conventional autoregressive moving average (ARMA) algorithm is applicable to perform modal identification, however, only for stationary-process vibration. The ergodicity postulate which has been conventionally employed for stationary processes is no longer valid in the case of nonstationary analysis. The objective of this paper is therefore to develop modal-identification techniques based on the nonstationary time series for linear systems subjected to nonstationary ambient excitation. Nonstationary ARMA model with time-varying parameters is considered because of its capability of resolving general nonstationary problems. The parameters of moving averaging (MA) model in the nonstationary time-series algorithm are treated as functions of time and may be represented by a linear combination of base functions and therefore can be used to solve the identification problem of time-varying parameters. Numerical simulations confirm the validity of the proposed modal-identification method from nonstationary ambient response data.

Sign in / Sign up

Export Citation Format

Share Document