• The Korean Journal of Applied Statistics
• /
• v.31 no.2
• /
• pp.205-216
• /
• 2018

This paper is concerned with volatility computations for high frequency time series. A threshold-asymmetric realized volatility (T-RV) is suggested to capture a leverage effect. The T-RV is compared with various conventional volatility computations including standard realized volatility, GARCH-type volatilities, historical volatility and exponentially weighted moving average volatility. High frequency KOSPI data are analyzed for illustration.

• Communications for Statistical Applications and Methods
• /
• v.31 no.1
• /
• pp.105-127
• /
• 2024

We propose recurrent neural network (RNN) methods for forecasting realized volatility (RV). The data are RVs of ten major stock price indices, four from the US, and six from the EU. Forecasts are made for relative ratio of adjacent RVs instead of the RV itself in order to avoid the out-of-scale issue. Forecasts of RV ratios distribution are first constructed from which those of RVs are computed which are shown to be better than forecasts constructed directly from RV. The apparent asymmetry of RV ratio is addressed by the Piecewise Min-max (PM) normalization. The serial dependence of the ratio data renders us to consider two architectures, long short-term memory (LSTM) and gated recurrent unit (GRU). The hyperparameters of LSTM and GRU are tuned by the nested cross validation. The RNN forecast with the PM normalization and ratio transformation is shown to outperform other forecasts by other RNN models and by benchmarking models of the AR model, the support vector machine (SVM), the deep neural network (DNN), and the convolutional neural network (CNN).

• The Korean Journal of Applied Statistics
• /
• v.35 no.1
• /
• pp.93-104
• /
• 2022

In forecasting realized volatility of the major US stock price indexes (S&P 500, Russell 2000, DJIA, Nasdaq 100), internet search volume reflecting investor's interests and implied volatility are used to improve forecast via a deep learning method of the LSTM. The LSTM method combined with search volume index produces better forecasts than existing standard methods of the vector autoregressive (VAR) and the vector error correction (VEC) models. It also beats the recently proposed vector error correction heterogeneous autoregressive (VECHAR) model which takes advantage of the cointegration relation between realized volatility and implied volatility.

• The Korean Journal of Applied Statistics
• /
• v.25 no.1
• /
• pp.15-27
• /
• 2012

We have studied the realized variance(RV) of intra-day returns and market microstructure noise based on high-frequency stock transaction data for the four largest companies in terms of market capitalization in the KOSPI. First, non-negligible biases are observed for the RV and for the bias-corrected realized variance($RV_{AC_1}$) which is constructed by adjusting RV for the first order autocorrelation in intra-day returns. Bias is more obvious for the RV and the $RV_{AC_1}$ when intra-day returns are sampled more frequently than every 2 minutes. Transaction Time Sampling(TTS) is shown to be better than Calendar Time Sampling(CTS) in terms of biases of the RV and the $RV_{AC_1}$ for the 4 companies. The analysis reveals that market microstructure noise is temporally dependent. Second, by using the Noise-to-Signal Ratio(NSR), we estimate sampling frequencies that are optimal in terms of the Mean Square Errors(MSE) of the RV and the $RV_{AC_1}$. The optimal sampling frequencies are around 200 for RV and is around 5000 for the $RV_{AC_1}$ for all the four stock prices. For the 6 hour transaction period of the Korean stock trading, these correspond to about 2 minutes and 6 seconds.

• 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.29 no.3
• /
• pp.505-512
• /
• 2016

The paper is concerned with high frequency financial time series. A weighted hybrid volatility is suggested to compute daily volatilities based on high frequency data. Various realized volatility (RV) computations are reviewed and the weights are chosen by minimizing the differences between the hybrid volatility and the realized volatility. A high frequency time series of KOSPI200 index is illustrated via QLIKE and Theil-U statistics.

• 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.

• Quantitative Bio-Science
• /
• v.37 no.2
• /
• pp.73-79
• /
• 2018

As high frequency (HF, for short) time series is now prevalent in the presence of real time big data, volatility computations based on traditional ARCH/GARCH models need to be further developed to suit the high frequency characteristics. This article reviews realized volatilities (RV) and multivariate GARCH (MGARCH) to deal with high frequency volatility computations. As a (functional) infinite dimensional models, the fARCH and fGARCH are introduced to accommodate ultra high frequency (UHF) volatilities. The fARCH and fGARCH models are developed in the recent literature by Hormann et al. [1] and Aue et al. [2], respectively, and our discussions are mainly based on these two key articles. Real data applications to domestic UHF financial time series are illustrated.

• The Korean Journal of Applied Statistics
• /
• v.29 no.1
• /
• pp.85-98
• /
• 2016

This study makes an empirical comparison of various realized volatilities (RVs) in terms of overnight returns. In financial asset markets, during overnight or holidays, no or few trading data are available causing a difficulty in computing RVs for a whole span of a day. A review will be made on several RVs reflecting overnight return variations. The comparison is made for forecast accuracies of several RVs for some financial assets: the US S&P500 index, the US NASDAQ index, the KOSPI (Korean Stock Price Index), and the foreign exchange rate of the Korea won relative to the US dollar. The RV of a day is compared with the square of the next day log-return, which is a proxy for the integrated volatility of the day. The comparison is made by investigating the Mean Absolute Error (MAE) and the Root Mean Square Error (RMSE). Statistical inference of MAE and RMSE is made by applying the model confidence set (MCS) approach and the Diebold-Mariano test. For the three index data, a specific RV emerges as the best one, which addresses overnight return variations by inflating daytime RV.

• The Korean Journal of Applied Statistics
• /
• v.30 no.5
• /
• pp.747-757
• /
• 2017

We investigate multivariate volatilities based on high frequency time series. The PCA (principal component analysis) method is employed to achieve a dimension reduction in multivariate volatility. Multivariate realized volatilities (RV) with various frequencies are calculated from high frequency data and "optimum" frequency is suggested using PCA. Specifically, RVs with various frequencies are compared with existing daily volatilities such as Cholesky, EWMA and BEKK after dimension reduction via PCA. An analysis of high frequency stock prices of KOSPI, Samsung Electronics and Hyundai motor company is illustrated.