• 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.

• 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).

• Journal of Information Technology Applications and Management
• /
• v.19 no.1
• /
• pp.1-12
• /
• 2012

Data mining is the process of searching and analyzing large quantities of data for finding out meaningful patterns and rules. Artificial Neural Network (ANN) is one of the tools of data mining which is becoming very popular in forecasting the future values. Some of the areas where it is used are banking, medicine, retailing and fraud detection. In finance, artificial neural network is used in various disciplines including stock market forecasting. In the stock market time series, due to high volatility, it is very important to choose a model which reads volatility and forecasts the future values considering volatility as one of the major attributes for forecasting. In this paper, an attempt is made to develop two models - one using feed forward back propagation Artificial Neural Network and the other using Autoregressive Conditional Heteroskedasticity (ARCH) technique for forecasting stock market returns. Various parameters which are considered for the design of optimal ANN model development are input and output data normalization, transfer function and neuron/s at input, hidden and output layers, number of hidden layers, values with respect to momentum, learning rate and error tolerance. Simulations have been done using prices of daily close of Sensex. Stock market returns are chosen as input data and output is the forecasted return. Simulations of the Model have been done using MATLAB$^{(R)}$ 6.1.0.450 and EViews 4.1. Convergence and performance of models have been evaluated on the basis of the simulation results. Performance evaluation is done on the basis of the errors calculated between the actual and predicted values.

• Communications for Statistical Applications and Methods
• /
• v.23 no.3
• /
• pp.203-213
• /
• 2016

We investigate volatility spillover aspects of realized volatilities (RVs) for the log returns of the Korea Composite Stock Price Index (KOSPI) and the Hang Seng Index (HSI) from 2009-2013. For all RVs, significant long memories and asymmetries are identified. For a model selection, we consider three commonly used time series models as well as three models that incorporate long memory and asymmetry. Taking into account of goodness-of-fit and forecasting ability, Leverage heteroskedastic autoregressive realized volatility (LHAR) model is selected for the given data. The LHAR model finds significant decompositions of the spillover effect from the HSI to the KOSPI into moderate negative daily spillover, positive weekly spillover and positive monthly spillover, and from the KOSPI to the HSI into substantial negative weekly spillover and positive monthly spillover. An interesting result from the analysis is that the daily volatility spillover from the HSI to the KOSPI is significant versus the insignificant daily volatility spillover of the KOSPI to HSI. The daily volatility in Hong Kong affects next day volatility in Korea but the daily volatility in Korea does not affect next day volatility in Hong Kong.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.2
• /
• pp.327-335
• /
• 2016

The People's Republic of China has vigorously been pursuing the internationalization of the Chinese Yuan or Renminbi after the financial crisis of 2008. In this view, an abrupt increase of use of the Chinese Yuan in the onshore and offshore markets are important milestones to be one of important currencies. One of the most frequently used methods to forecast volatility is GARCH model. Since a prediction error of the GARCH model has been reported quite high, a lot of efforts have been made to improve forecasting capability of the GARCH model. In this paper, we have proposed MLP-GARCH and a DL-GARCH by employing Artificial Neural Network to the GARCH. In an application to forecasting Chinese Yuan volatility, we have successfully shown their overall outperformance in forecasting over the GARCH.

• KDI Journal of Economic Policy
• /
• v.44 no.4
• /
• pp.79-108
• /
• 2022

This study examines the impact of sentiment shock, which is defined as a stochastic innovation to the Housing Market Confidence Index (HMCI) that is orthogonal to past housing price changes, on aggregate housing price changes and housing price volatility. This paper documents empirical evidence that sentiment shock has a statistically significant relationship with Korea's aggregate housing price changes. Specifically, the key findings show that an increase in sentiment shock predicts a rise in the aggregate housing price and a drop in its volatility at the national level. For the Seoul Metropolitan Region (SMR), this study also suggests that sentiment shock is positively associated with one-month-ahead aggregate housing price changes, whereas an increase in sentiment volatility tends to increase housing price volatility as well. In addition, the out-of-sample forecasting exercises conducted here reveal that the prediction model endowed with sentiment shock and sentiment volatility outperforms other competing prediction models.

• International Commerce and Information Review
• /
• v.10 no.1
• /
• pp.395-414
• /
• 2008

The objective of this study is to introduce the concepts and theories of conditional heteroscedastic volatility models and the news impact curves and apply them to the Korea inbound tourism market. Three volatility models were introduced and used to estimate the conditional volatility of monthly arrivals of inbound tourists into Korea and news impact curves according to the three models. Results of this study are as follows. As the proportion of American tourists occupied a large amount of Korea inbound tourism market, the markets' forecasting is very important. The news impact curves which used EGARCH model (1,1) and TGARCH model(1,1), with data on these tourists to Korea showed an asymmetry effect of volatility. It was common that bad news means that it was estimated more sensitively than good news. From these results, we will notice that American tourists who visited Korea only for tourism are affected by good news. The result suggests that the Korea government and tourism industry should pay more attention to changes in the tourism environment following bad news because conditional volatility increases more when a negative shock occurs than when a positive shock occurs.

• The Journal of Asian Finance, Economics and Business
• /
• v.8 no.7
• /
• pp.11-19
• /
• 2021

The study aims to investigate the impact of seasonality in Gulf Cooperation Council (GCC) countries' share market during the month of Ramadan. It helps in finding the opportunities for stock market investors to earn abnormal (returns) gain by investing during Ramadan in GCC stock markets. This study uses stock returns data of GCC countries (Saudi Arabia, Bahrain, Qatar, Kuwait, Dubai, and UAE) from January 2004 to November 2019. Stock prices indexes of GCC stock markets have been obtained from Datastream. The ARCH-GARCH model is used to study the impact of the Ramadan month on the return and volatility of the stock market in GCC countries. The results showed that the Ramadan month has a significant impact on share market prices in Saudi Arabia and the United Arab Emirates. However, Ramadan has an insignificant impact on share market prices in Bahrain and Oman. The study found no evidence of serial correlational between residuals in Kuwait; meaning that stock return was not dependent on the prior stock returns in Kuwait, therefore, we cannot go for forecasting. The ARCH-LM test statistic for Qatar does not fulfill the requirement of a good regression model; therefore, we cannot go for forecasting or testing the hypothesis of Qatar.

• The Korean Journal of Applied Statistics
• /
• v.32 no.1
• /
• pp.69-81
• /
• 2019

We consider bivariate long range dependent (LRD) time series forecasting using a deep learning method. A long short-term memory (LSTM) network well-suited to time series data is applied to forecast bivariate time series; in addition, we compare the forecasting performance with bivariate fractional autoregressive integrated moving average (FARIMA) models. Out-of-sample forecasting errors are compared with various performance measures for functional MRI (fMRI) data and daily realized volatility data. The results show a subtle difference in the predicted values of the FIVARMA model and VARFIMA model. LSTM is computationally demanding due to hyper-parameter selection, but is more stable and the forecasting performance is competitively good to that of parametric long range dependent time series models.

• International journal of advanced smart convergence
• /
• v.12 no.4
• /
• pp.75-87
• /
• 2023

We designed to employ an Artificial Intelligence learning model to predict real estate prices and determine the reasons behind their changes, with the goal of using the results as a guide for policy. Numerous studies have already been conducted in an effort to develop a real estate price prediction model. The price prediction power of conventional time series analysis techniques (such as the widely-used ARIMA and VAR models for univariate time series analysis) and the more recently-discussed LSTM techniques is compared and analyzed in this study in order to forecast real estate prices. There is currently a period of rising volatility in the real estate market as a result of both internal and external factors. Predicting the movement of real estate values during times of heightened volatility is more challenging than it is during times of persistent general trends. According to the real estate market cycle, this study focuses on the three times of extreme volatility. It was established that the LSTM, VAR, and ARIMA models have strong predictive capacity by successfully forecasting the trading price index during a period of unusually high volatility. We explores potential synergies between the hybrid artificial intelligence learning model and the conventional statistical prediction model.