|
on Econometric Time Series |
| By: | Chang, Jinyuan; Zhang, Henry; Yang, Lin; Yao, Qiwei |
| Abstract: | We consider to model matrix time series based on a tensor canonical polyadic (CP)-decomposition. Instead of using an iterative algorithm which is the standard practice for estimating CP-decompositions, we propose a new and one-pass estimation procedure based on a generalized eigenanalysis constructed from the serial dependence structure of the underlying process. To overcome the intricacy of solving a rank-reduced generalized eigenequation, we propose a further refined approach which projects it into a lower-dimensional full-ranked eigenequation. This refined method can significantly improve the finite-sample performance. We show that all the component coefficient vectors in the CP-decomposition can be estimated consistently. The proposed model and the estimation method are also illustrated with both simulated and real data, showing effective dimension-reduction in modelling and forecasting matrix time series. |
| Keywords: | dimension-reduction; generalized eigenanalysis; tensor CP-decomposition; matrix time series |
| JEL: | C1 |
| Date: | 2023–02–01 |
| URL: | http://d.repec.org/n?u=RePEc:ehl:lserod:117644&r=ets |
| By: | Chaoyi Chen; Yiguo Sun; Yao Rao |
| Abstract: | We propose several threshold mixed data sampling (TMIDAS) autoregressive models to forecast the Canadian inflation rate using predictors observed at different frequencies. These models take two low-frequency variables and a high-frequency index as a threshold variable. We compare our TMIDAS models to commonly used benchmark models, evaluating their in-sample and out-of-sample forecasts. Our results demonstrate the good forecasting performance of the TMIDAS models. Particularly, the in-sample results highlight that the TMIDAS model using the high-frequency index as the threshold variable outperforms other models. Through unconditional superior predictive ability (USPA) and conditional superior predictive ability (CSPA) tests for out-of-sample evaluation, we find that no single model consistently outperforms the others, although at least one of our TMIDAS models remains competitive in most cases |
| Keywords: | Forecasting; High-frequency index; Mixed data sampling; Superiority predictive ability test; Threshold regression |
| JEL: | C24 C53 |
| Date: | 2023–07 |
| URL: | http://d.repec.org/n?u=RePEc:liv:livedp:202314&r=ets |
| By: | E. Otranto |
| Abstract: | Modern approaches to financial time series aim to model in a multivariate framework the volatility of different indices or assets, which could influence each other, creating spillover effects. Furthermore, the integration of financial markets provides a similar dynamics (co-movement). We propose a new model for volatility vectors, belonging to the family of Multiplicative Error Models, which incorporates spillover and co-movement effects. By adopting an appropriate parameterization, it is possible to estimate this model even for high dimensional vectors of volatility. To reduce the number of unknown coefficients, we propose a 3-step model-based clustering procedure. The proposed model is applied to a set of seventeen world financial indices, providing a useful interpretation of spillover effects and co- movements. Furthermore, the proposed parameterization is compared with two alternatives, showing significantly better performance. |
| Keywords: | high-dimensional time series;vector of volatility;multiplicative factors;model-based clustering;high-low range |
| Date: | 2024 |
| URL: | http://d.repec.org/n?u=RePEc:cns:cnscwp:202404&r=ets |