• Communications for Statistical Applications and Methods
• /
• v.25 no.6
• /
• pp.659-671
• /
• 2018

In this study, we consider the extension of the heterogeneous autoregressive (HAR) model for realized volatility by incorporating a neural network (NN) structure. Since HAR is a linear model, we expect that adding a neural network term would explain the delicate nonlinearity of the realized volatility. Three neural network-based HAR models, namely HAR-NN, $HAR({\infty})-NN$, and HAR-AR(22)-NN are considered with performance measured by evaluating out-of-sample forecasting errors. The results of the study show that HAR-NN provides a slightly wider interval than traditional HAR as well as shows more peaks and valleys on the turning points. It implies that the HAR-NN model can capture sharper changes due to higher volatility than the traditional HAR model. The HAR-NN model for prediction interval is therefore recommended to account for higher volatility in the stock market. An empirical analysis on the multinational realized volatility of stock indexes shows that the HAR-NN that adds daily, weekly, and monthly volatility averages to the neural network model exhibits the best performance.

• The Korean Journal of Applied Statistics
• /
• v.36 no.4
• /
• pp.295-307
• /
• 2023

The heterogeneous autoregressive (HAR) model is a simple linear model that is commonly used to explain long memory in the realized volatility. However, as realized volatility has more complicated features such as conditional heteroscedasticity, leverage effect, and volatility clustering, it is necessary to extend the simple HAR model. Therefore, to better incorporate the stylized facts, we propose a threshold HAR model with GARCH errors, namely the THAR-GARCH model. That is, the THAR-GARCH model is a nonlinear model whose coefficients vary according to a threshold value, and the conditional heteroscedasticity is explained through the GARCH errors. Model parameters are estimated using an iterative weighted least squares estimation method. Our simulation study supports the consistency of the iterative estimation method. In addition, we show that the proposed THAR-GARCH model has better forecasting power by applying to the realized volatility of major 21 stock indices around the world.

• The Korean Journal of Applied Statistics
• /
• v.34 no.2
• /
• pp.239-254
• /
• 2021

This paper studies time series analysis with estimation and forecasting for Korean COVID-19 confirmed cases, based on the approach of a heterogeneous autoregressive (HAR) model with two-piece t (TP-T) distributed errors. We consider HAR-TP-T time series models and suggest a step-by-step method to estimate HAR coefficients as well as TP-T distribution parameters. In our proposed step-by-step estimation, the ordinary least squares method is utilized to estimate the HAR coefficients while the maximum likelihood estimation (MLE) method is adopted to estimate the TP-T error parameters. A simulation study on the step-by-step method is conducted and it shows a good performance. For the empirical analysis on the Korean COVID-19 confirmed cases, estimates in the HAR-TP-T models of order p = 2, 3, 4 are computed along with a couple of selected lags, which include the optimal lags chosen by minimizing the mean squares errors of the models. The estimation results by our proposed method and the solely MLE are compared with some criteria rules. Our proposed step-by-step method outperforms the MLE in two aspects: mean squares error of the HAR model and mean squares difference between the TP-T residuals and their densities. Moreover, forecasting for the Korean COVID-19 confirmed cases is discussed with the optimally selected HAR-TP-T model. Mean absolute percentage error of one-step ahead out-of-sample forecasts is evaluated as 0.0953% in the proposed model. We conclude that our proposed HAR-TP-T time series model with optimally selected lags and its step-by-step estimation provide an accurate forecasting performance for the Korean COVID-19 confirmed cases.

• The Korean Journal of Applied Statistics
• /
• v.30 no.3
• /
• pp.323-334
• /
• 2017

We note that the dynamics of realized volatilities (RVs) are near the boundary between stationarity and non-stationarity because RVs have persistent long-memory and are often subject to fairly large outlying values. To forecast realized volatility, we consider a new method that adaptively use models with and without unit root according to the abnormality of observed RV: heterogeneous autoregressive (HAR) model and the Integrated HAR (IHAR) model. The resulting method is called the IHAR-O-HAR method. In an out-of-sample forecast comparison for the realized volatility datasets of the 3 major indexes of the S&P 500, the NASDAQ, and the Nikkei 225, the new IHAR-O-HAR method is shown superior to the existing HAR and IHAR method.

• The Korean Journal of Applied Statistics
• /
• v.29 no.7
• /
• pp.1213-1229
• /
• 2016

Cho and Shin (2016) recently demonstrated that an integrated HAR model has a forecast advantage over the HAR model of Corsi (2009). Recalling that realized volatilities of financial assets have asymmetries, we add a leverage term to the integrated HAR model, yielding the LIHAR model. Out-of-sample forecast comparisons show superiority of the LIHAR model over the HAR and IHAR models. The comparison was made for all the 20 realized volatilities in the Oxford-Man Realized Library focusing specially on the DJIA, the S&P 500, the Russell 2000, and the KOSPI. Analysis of the realized volatility data sets reveal apparent long-memory and asymmetry. The LIHAR model takes advantage of the long-memory and asymmetry and produces better forecasts than the HAR, IHAR, LHAR models.

• Journal of Korea Multimedia Society
• /
• v.25 no.8
• /
• pp.1049-1058
• /
• 2022

HAR(Human Action Recognition) such as anomaly and object detection has become a trend in research field(s) that focus on utilizing Artificial Intelligence (AI) methods to analyze patterns of human action in crime-ridden area(s), media services, and industrial facilities. Especially, in real-time system(s) using video streaming data, HAR has become a more important AI-based research field in application development and many different research fields using HAR have currently been developed and improved. In this paper, we propose and analyze a deep-learning-based HAR that provides more efficient scheme(s) using an intelligent AI models, such system can be applied to media services using RGB video streaming data usage without feature extraction pre-processing. For the method, we adopt Slowfast based on the Deep Neural Network(DNN) model under an open dataset(HMDB-51 or UCF101) for improvement in prediction accuracy.

• Environmental and Resource Economics Review
• /
• v.30 no.4
• /
• pp.535-555
• /
• 2021

This study empirically analyzes the effects of international oil prices on domestic stock market volatility. The data used for the analysis are 10-minute high-frequency data of the KOSPI index and WTI futures price from January 2, 2015, to July 30, 2021. For using the high-frequency data, a heterogeneous autoregression (HAR) model is employed. The analysis model utilizes the advantages of high frequency data to observe the impact of international oil prices through realized volatility, realized skewness, and kurtosis as well as oil price return. In the estimation, the Box-Cox transformation is applied in consideration of the distribution of realized volatility with high skewness. As a result, it finds that the daily return fluctuation of the WTI price has a statistically significant positive (+) effect on the volatility of the KOSPI return. However, the volatility, skewness, and kurtosis of the WTI return do not appear to affect the volatility of the KOSPI return. This result is believed to be because the volatility of the KOSPI return reflects the daily change in the WTI return, but does not reflect the intraday trading behavior of investors.

• ETRI Journal
• /
• v.33 no.4
• /
• pp.569-579
• /
• 2011

This paper presents a novel approach for human activity recognition (HAR) using the joint angles from a 3D model of a human body. Unlike conventional approaches in which the joint angles are computed from inverse kinematic analysis of the optical marker positions captured with multiple cameras, our approach utilizes the body joint angles estimated directly from time-series activity images acquired with a single stereo camera by co-registering a 3D body model to the stereo information. The estimated joint-angle features are then mapped into codewords to generate discrete symbols for a hidden Markov model (HMM) of each activity. With these symbols, each activity is trained through the HMM, and later, all the trained HMMs are used for activity recognition. The performance of our joint-angle-based HAR has been compared to that of a conventional binary and depth silhouette-based HAR, producing significantly better results in the recognition rate, especially for the activities that are not discernible with the conventional approaches.

• Communications for Statistical Applications and Methods
• /
• v.24 no.4
• /
• pp.367-382
• /
• 2017

We consider an infinite-order long-memory heterogeneous autoregressive (HAR) model, which is motivated by a long-memory property of realized volatilities (RVs), as an extension of the finite order HAR-RV model. We develop bootstrap tests for structural mean or variance changes in the infinite-order HAR model via stationary bootstrapping. A functional central limit theorem is proved for stationary bootstrap sample, which enables us to develop stationary bootstrap cumulative sum (CUSUM) tests: a bootstrap test for mean break and a bootstrap test for variance break. Consistencies of the bootstrap null distributions of the CUSUM tests are proved. Consistencies of the bootstrap CUSUM tests are also proved under alternative hypotheses of mean or variance changes. A Monte-Carlo simulation shows that stationary bootstrapping improves the sizes of existing tests.

• The Korean Journal of Applied Statistics
• /
• v.35 no.1
• /
• pp.49-62
• /
• 2022

Realized volatility is well known to have long memory, strong association with other global financial markets and interdependences among macroeconomic indices such as exchange rate, oil price and interest rates. This paper proposes autoencoder factor-augmented heterogeneous autoregressive (AE-FAHAR) model for realized volatility forecasting. AE-FAHAR incorporates long memory using HAR structure, and exogenous variables into few factors summarized by autoencoder. Autoencoder requires intensive calculation due to its nonlinear structure, however, it is more suitable to summarize complex, possibly nonstationary high-dimensional time series. Our AE-FAHAR model is shown to have smaller out-of-sample forecasting error in empirical analysis. We also discuss pre-training, ensemble in autoencoder to reduce computational cost and estimation errors.