Modelling Multivariate Time Series
نویسنده
چکیده
Multivariate time series (MTS) data are widely available in di erent elds including medicine, nance, science and engineering. Modelling MTS data e ectively is important for many decision-making activities. In this paper, we will describe some of our e orts in modelling these data for numerous tasks such as outlier analysis, forecasting and explanation. Through the analysis of various kinds of MTS data to achieve those tasks, we hope to trigger discussions in the workshop about what \I" could mean in \IDA" for this type of application.
منابع مشابه
Time series forecasting of Bitcoin price based on ARIMA and machine learning approaches
Bitcoin as the current leader in cryptocurrencies is a new asset class receiving significant attention in the financial and investment community and presents an interesting time series prediction problem. In this paper, some forecasting models based on classical like ARIMA and machine learning approaches including Kriging, Artificial Neural Network (ANN), Bayesian method, Support Vector Machine...
متن کاملVolatility Modelling of Multivariate Financial Time Series by Using ICA-GARCH Models
Volatility modelling of asset returns is an important aspect for many financial applications, e.g., option pricing and risk management. GARCH models are usually used to model the volatility processes of financial time series. However, multivariate GARCH modelling of volatilities is still a challenge due to the complexity of parameters estimation. To solve this problem, we suggest using Independ...
متن کاملGraphical modelling for multivariate time series
Graphical models for multivariate time series is a concept extended by Dahlhaus (2000) from a random vector to a time series. We propose a test statistic to identify a graphical model for multivariate time series with the Kullback-Leibler distance between two spectral density matrices characterized by graphical models. Asymptotic null distribution is derived to be normal with the mean and varia...
متن کاملThe Modelling of Short High-Dimensional Multivariate Time Series
In bio-medical domains there are many applications involving the modelling of multivariate time series (MTS) data. One area that has been largely overlooked so far is the particular type of time series where the dataset consists of a large number of variables but with a small number of observations. This thesis presents a methodology for the modelling of this type of data and introduces a novel...
متن کاملa Comparison Study Between the Joint Probability Approach and Time Series Rainfall Modelling in Coastal Detention Pond Analysis (RESEARCH NOTE)
In tidally affected coastal catchments detention pond should be provided to store flood surface water. A comparison between the full simulation approach based on the joint probability method and time series rainfall modeling via the annual maximum of pond level was undertaken to investigate the assumptions of independence between variables that are necessary in the joint probability method. The...
متن کامل