Mixed-frequency SV model for stock volatility and macroeconomics

Water Resources Research

Song

Zhang

et al. 2023

The analysis of hydrological series is important for water resource management and allocation, as well as energy and food security in the face of climate change (Arriagada et al., 2019;Tencaliec et al., 2015). As one of the most important indicators in hydrological analysis, providing an important source of hydrological information for water resources engineering applications (Birsan, 2015;Shiau & Hsu, 2016). However, streamflow series have strong dynamic spatiotemporal characteristics, making accurate simulation and prediction difficult. This necessitates the use of highly complex tools to model the sophisticated features of streamflow series.In general, hydrological models are classified into two types: physical models and statistical models. Specifically, physical models are complicated because they take into account a large number of variables that influence hydrological processes. Statistical models simulate hydrological changes using observational data collected beforehand. Owing to their simple implementation, statistical models are easier to use in practice (Modarres & Ouarda, 2013b). Data-driven technologies, such as artificial neural networks, support vector machines, wavelet transforms, and time series analysis (TSA) models, have been receiving increasing attention in the field of hydrology (Hadi et al., 2020;Khan et al., 2020;Parisouj et al., 2020). In the application of the aforementioned data-driven models, the development of TSA models and new methods have consistently been the main theme in hydrological modeling (Fathian, Fard, et al., 2019;Fathian et al., 2018).In particular, linear time series models have been widely used to simulate hydroclimatic time series (Hu et al., 2020;H. Wang et al., 2021). Traditional time series models, such as the autoregressive integrated moving average (ARIMA) model, have primarily been developed for simulating the first-moment behavior of time series, but they ignore the second-moment information (volatility), leading to insufficient model fitting results (Fathian, Fard, et al., 2019). In reality, hydrological time series have strong volatility associated with environmental change.

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Stochastic Volatility Modeling of Daily Streamflow Time Series

Water Resources Research

Song

Zhang

et al. 2023

“…In considering macroeconomic variables and mixed-frequency data forecasting models, the generalized autoregressive conditional heteroskedasticity-MIDAS (GARCH-MIDAS) model was proposed by Engle et al. (2013), while Shang and Zheng (2021) proposed the stochastic volatility-MIDAS (SV-MIDAS) model. In other types of MIDAS models, Lu et al.…”

Section: Literature Reviewmentioning

confidence: 99%

Mixed-frequency data-driven forecasting the important economies' performance in a smart city: a novel RUMIDAS-SVR model

Zhang

et al. 2022

IMDS

PurposeThe purpose of this study is to forecast the development performance of important economies in a smart city using mixed-frequency data.Design/methodology/approachThis study introduces reverse unrestricted mixed-data sampling (RUMIDAS) to support vector regression (SVR) to develop a novel RUMIDAS-SVR model. The RUMIDAS-SVR model was estimated using a quadratic programming problem. The authors then use the novel RUMIDAS-SVR model to forecast the development performance of all high-tech listed companies, an important sector of the economy reflecting the potential and dynamism of urban economic development in Shanghai using the mixed-frequency consumer price index (CPI) producer price index (PPI), and consumer confidence index (CCI) as predictors.FindingsThe empirical results show that the established RUMIDAS-SVR is superior to the competing models with regard to mean absolute error (MAE) and root-mean-squared error (RMSE) and multi-source macroeconomic predictors contribute to the development performance forecast of important economies.Practical implicationsSmart city policy makers should create a favourable macroeconomic environment, such as controlling inflation or stabilising prices for companies within the city, and companies within the important city economic sectors should take initiative to shoulder their responsibility to support the construction of the smart city.Originality/valueThis study contributes to smart city monitoring by proposing and developing a new model, RUMIDAS-SVR, to help the construction of smart cities. It also empirically provides strategic insights for smart city stakeholders.

Economic Research-Ekonomska Istraživanja

“…In the existing literature, stock volatility is studied based on a single frequency, while ignoring the impact of different frequency information on volatility. Presently, the mixed-frequency technique of mixed data sampling (MIDAS) model is used to predict volatility (see, e.g., Andreou, 2016;Mei et al, 2020;Shang & Zheng, 2021). Some scholars have also proposed the mixed frequency method, different from MIDAS mechanism.…”

Section: Literature Reviewmentioning

confidence: 99%

Modelling returns volatility: mixed-frequency model based on momentum of predictability

Chen

Shang

2022

The estimation and prediction of financial asset volatility are important in terms of theoretical and practical applications. Considering that low-frequency and high-frequency information plays an important role in volatility prediction, this article proposes a mixed-frequency model based on the momentum of predictability (MF-MoP). To illustrate the advantages of the proposed model, comparative research is conducted on the prediction accuracy of volatility among the GARCH model, the Realized GARCH model and the MF-MoP model, by the loss function and MCS test. The empirical results show that the MF-MoP model has higher prediction accuracy than the other two models; especially based on skewed-t distribution, the MF-MoP significantly outperforms the competing models. Moreover, the MF-MoP model can improve the forecasting of volatility, regardless of different lookback periods (including 1, 3, 6 and 9 days), different data (including the CSI 300 index, the N225 index and the KS11 index), and realized measures (including RV, RRV and MedRV), indicating that the model is robust.