Non-linear and Non-Gaussian Time Series Modeling
The research in this area is concentrated on understanding complex, non-linear stochastic time series, particularly in the context of finance.
Researchers focus on statistical methods for these intricate systems and apply them to financial scenarios. Various versions of GARCH-type volatility models are explored to analyze statistical aspects of conditional heteroscedastic models, commonly used in modeling financial volatility. Specifically, non-Gaussian and non-linear multivariate time series are employed to examine how different scales of macroeconomic factors influence managing financial risks and impact correlations among financial assets concerning market variables. Additionally, the work extends to using latent factor time series models to study historical economic data and employing wavelet analysis in statistics to analyze long-memory processes. These diverse applications showcase the versatility of their research in the realm of finance and statistical analysis.