scholarly journals Time Series Analysis of Electricity Spot Price with AR-GARCH Model and its Application

2007 ◽  
Vol 127 (1) ◽  
pp. 61-68 ◽  
Author(s):  
Yasuyuki Itoh ◽  
Yoshiki Murakami ◽  
Takenori Kobayashi
Keyword(s):  
Time Series ◽  
Time Series Analysis ◽  
Garch Model ◽  
Spot Price ◽  
Series Analysis ◽  
Electricity Spot Price

Time Series Analysis for the COMEX Gold Spot Price Forecasting by Using NARX DE/SVR and DE/GPR Techniques

2021 ◽  
pp. 145-154
Author(s):  
Esperanza García-Gonzalo ◽  
Paulino José García Nieto ◽  
Gregorio Fidalgo Valverde ◽  
Pedro Riesgo Fernández ◽  
Fernando Sánchez Lasheras
Keyword(s):  
Time Series ◽  
Time Series Analysis ◽  
Spot Price ◽  
Price Forecasting ◽  
Series Analysis

scholarly journals PEMODELAN VOLATILITAS RETURN PORTOFOLIO SAHAM MENGGUNAKAN FEED FORWARD NERURAL NETWORK (Studi Kasus :PT Bumi Serpong Damai Tbk. Dan PT H.M Sampoerna Tbk.)

2018 ◽  
Vol 7 (2) ◽  
pp. 200-211
Author(s):  
Rizki Pradipto Widyantomo ◽  
Abdul Hoyyi ◽  
Tatik Widiharih
Keyword(s):  
Neural Network ◽  
Time Series ◽  
Time Series Analysis ◽  
Network Model ◽  
Neural Network Model ◽  
Arima Model ◽  
Garch Model ◽  
Residual Variance ◽  
Series Analysis ◽  
Input Variables

Time series analysis is an analysis used to predict a time-observed data, one of which is the ARIMA model. ARIMA model assumes a constant residual variance (homogeneous). While financial data usually produce ARIMA model with variance error that is not constant. If the assumption of homogeneity of the residual variance is not met, then the method that can be used is ARCH or GARCH model. Another method that can be used on the data assuming the homogeneity of the variance error is not met is the Neural Network model. In this model we use Neural Network model with variance and residual as the input variables that obtained from ARCH / GARCH model. The data used are BSDE and HMSP asset portfolio returns from November 14, 2016 to January 18, 2018. In this study the selected input variables are from ARIMA (1.0.1) GARCH (1,1) model. The best Neural Network model obtained is Neural Network model with 10 hidden layers with MSE value 6.58 x10-10 with model train evaluation which is MAPE value 1.14441%.Keywords: Time series Analysis, ARCH / GARCH, Neural Network, Return.

Dynamical Peculiarities of Concatenational Changes in Functional Cardiovascular Parameters

2018 ◽  
Vol 3 (82) ◽  
Author(s):  
Eurelija Venskaitytė ◽  
Jonas Poderys ◽  
Tadas Česnaitis
Keyword(s):  
Time Series ◽  
Time Series Analysis ◽  
Functional State ◽  
The Body ◽  
Functional Changes ◽  
High Intensity Training ◽  
Series Analysis ◽  
Greco Roman ◽  
St Segment ◽  
Standard Lead

Research  background  and  hypothesis.  Traditional  time  series  analysis  techniques,  which  are  also  used  for the analysis of cardiovascular signals, do not reveal the relationship between the  changes in the indices recorded associated with the multiscale and chaotic structure of the tested object, which allows establishing short-and long-term structural and functional changes.Research aim was to reveal the dynamical peculiarities of interactions of cardiovascular system indices while evaluating the functional state of track-and-field athletes and Greco-Roman wrestlers.Research methods. Twenty two subjects participated in the study, their average age of 23.5 ± 1.7 years. During the study standard 12 lead electrocardiograms (ECG) were recorded. The following ECG parameters were used in the study: duration of RR interval taken from the II standard lead, duration of QRS complex, duration of JT interval and amplitude of ST segment taken from the V standard lead.Research  results.  Significant  differences  were  found  between  inter-parametric  connections  of  ST  segment amplitude and JT interval duration at the pre and post-training testing. Observed changes at different hierarchical levels of the body systems revealed inadequate cardiac metabolic processes, leading to changes in the metabolic rate of the myocardium and reflected in the dynamics of all investigated interactions.Discussion and conclusions. It has been found that peculiarities of the interactions of ECG indices interactions show the exposure of the  functional changes in the body at the onset of the workload. The alterations of the functional state of the body and the signs of fatigue, after athletes performed two high intensity training sessions per day, can be assessed using the approach of the evaluation of interactions between functional variables. Therefore the evaluation of the interactions of physiological signals by using time series analysis methods is suitable for the observation of these processes and the functional state of the body.Keywords: electrocardiogram, time series, functional state.

Sign in / Sign up

Export Citation Format

Share Document