• Journal of Information Processing Systems
• /
• v.15 no.5
• /
• pp.1201-1210
• /
• 2019

Stock price is characterized as being mutable, non-linear and stochastic. These key characteristics are known to have a direct influence on the stock markets globally. Given that the stock price data often contain both linear and non-linear patterns, no single model can be adequate in modelling and predicting time series data. The autoregressive integrated moving average (ARIMA) model cannot deal with non-linear relationships, however, it provides an accurate and effective way to process autocorrelation and non-stationary data in time series forecasting. On the other hand, the neural network provides an effective prediction of non-linear sequences. As a result, in this study, we used a hybrid ARIMA and neural network model to forecast the monthly closing price of the Shanghai composite index and Shenzhen component index.

• Journal of Distribution Science
• /
• v.22 no.9
• /
• pp.49-63
• /
• 2024

Purpose: This study examines how psychological heuristics influence stock price dynamics in Korea's distribution industry after significant price shocks. Research Design, Data, and Methodology: The study analyzes daily stock price movements exceeding 10% for Korean distribution companies from 1993 to 2022. It establishes anchoring heuristic reference points, including the 52-week high and low, and segments the sample based on company size and volatility. Results: We analyzed a sample previously studied by Lee et al. (2023). Our findings indicate that when a stock experiences a positive (negative) price shock near its 52-week high (or lowest price), investors in large (small) companies exhibit an optimism (pessimism) bias. This leads to overreactions and subsequent stock price reversals after the event date. Conversely, when a stock encounters a negative (positive) price shock near its 52-week high (or lowest price), investorstend to underreact due to anchoring heuristics. Thisresultsin a drift effect on the stock price after the event day. Notably, investor behavior around 52-week highs or lows directly impacts their heuristic behavior related to those price points. Conclusions: This paper uniquely examines behavioral biases among distribution-related stock investors in Korea, shedding light on stock price reversal and drift effects.

• The Journal of Bigdata
• /
• v.4 no.2
• /
• pp.1-12
• /
• 2019

As Artificial Intelligence (AI) technology develops, it is applied to various fields such as image, voice, and text. AI has shown fine results in certain areas. Researchers have tried to predict the stock market by utilizing artificial intelligence as well. Predicting the stock market is known as one of the difficult problems since the stock market is affected by various factors such as economy and politics. In the field of AI, there are attempts to predict the ups and downs of stock price by studying stock price patterns using various machine learning techniques. This study suggest a way of predicting stock price patterns based on the Convolutional Neural Network(CNN) among machine learning techniques. CNN uses neural networks to classify images by extracting features from images through convolutional layers. Therefore, this study tries to classify candlestick images made by stock data in order to predict patterns. This study has two objectives. The first one referred as Case 1 is to predict the patterns with the images made by the same-day stock price data. The second one referred as Case 2 is to predict the next day stock price patterns with the images produced by the daily stock price data. In Case 1, data augmentation methods - random modification and Gaussian noise - are applied to generate more training data, and the generated images are put into the model to fit. Given that deep learning requires a large amount of data, this study suggests a method of data augmentation for candlestick images. Also, this study compares the accuracies of the images with Gaussian noise and different classification problems. All data in this study is collected through OpenAPI provided by DaiShin Securities. Case 1 has five different labels depending on patterns. The patterns are up with up closing, up with down closing, down with up closing, down with down closing, and staying. The images in Case 1 are created by removing the last candle(-1candle), the last two candles(-2candles), and the last three candles(-3candles) from 60 minutes, 30 minutes, 10 minutes, and 5 minutes candle charts. 60 minutes candle chart means one candle in the image has 60 minutes of information containing an open price, high price, low price, close price. Case 2 has two labels that are up and down. This study for Case 2 has generated for 60 minutes, 30 minutes, 10 minutes, and 5minutes candle charts without removing any candle. Considering the stock data, moving the candles in the images is suggested, instead of existing data augmentation techniques. How much the candles are moved is defined as the modified value. The average difference of closing prices between candles was 0.0029. Therefore, in this study, 0.003, 0.002, 0.001, 0.00025 are used for the modified value. The number of images was doubled after data augmentation. When it comes to Gaussian Noise, the mean value was 0, and the value of variance was 0.01. For both Case 1 and Case 2, the model is based on VGG-Net16 that has 16 layers. As a result, 10 minutes -1candle showed the best accuracy among 60 minutes, 30 minutes, 10 minutes, 5minutes candle charts. Thus, 10 minutes images were utilized for the rest of the experiment in Case 1. The three candles removed from the images were selected for data augmentation and application of Gaussian noise. 10 minutes -3candle resulted in 79.72% accuracy. The accuracy of the images with 0.00025 modified value and 100% changed candles was 79.92%. Applying Gaussian noise helped the accuracy to be 80.98%. According to the outcomes of Case 2, 60minutes candle charts could predict patterns of tomorrow by 82.60%. To sum up, this study is expected to contribute to further studies on the prediction of stock price patterns using images. This research provides a possible method for data augmentation of stock data.

• Journal of Distribution Science
• /
• v.21 no.9
• /
• pp.103-114
• /
• 2023

Purpose: This paper aims to assess psychological heuristics' effectiveness on cumulative returns after significant stock price changes. Specifically, it compares availability and anchoring heuristics' empirical validity due to conflicting stock return predictions. Research Design, Data, and Methodology: This paper analyzes stock price changes of Korean distribution industry stocks in the KOSPI market from January 2004 to July 2022, where daily fluctuations exceed 10%. It evaluates availability heuristics using daily KOSPI index changes and tests anchoring heuristics using 52-week high and low stock prices as reference points. Results: As a result of the empirical analysis, stock price reversals did not consistently appear alongside changes in the daily KOSPI index. By contrast, stock price drifts consistently appeared around the 52-week highest stock price and 52-week lowest stock price. The result of the multiple regression analysis which controlled for both company-specific and event-specific variables supported the anchoring heuristics. Conclusions: For stocks related to the Korean distribution industry in the KOSPI market, the anchoring heuristics theory provides a consistent explanation for stock returns after large-scale stock price fluctuations that initially appear to be random movements.

• The Journal of Asian Finance, Economics and Business
• /
• v.7 no.8
• /
• pp.143-150
• /
• 2020

This study aims to investigate the effect of oil price and exchange rate on the two Vietnamese stock market indices: VN index and HXN index. This study uses the daily data from August 1^st 2000 to October 25^th 2019 of the two Vietnamese stock indices: VN index and HNX index, the two oil price indices: BRENT and WTI, and the two exchange rates: US dollar to Vietnamese dong and Euro to Vietnamese dong. Due to the presence of heteroskedasticity in our data, we use GARCH (1,1) regression model to perform our analysis. Our findings show that the oil price has a significant positive effect on the two Vietnamese stock market indices. In terms of the stock index volatility, both the VN index and HNX index volatilities are negatively impacted by the return of oil price. While the conclusion about the impact of oil price remained consistent through all three robustness tests, the effect of exchange rate on Vietnamese stock market indices is not consistent. We find thatchanges of the USD/VND exchange rate significantly impact the return and volatility of HNX index only in GARCH (1,1) setting. Our analysis also survives a number of robustness tests.

• Management Science and Financial Engineering
• /
• v.15 no.2
• /
• pp.101-118
• /
• 2009

In quantitative contexts, data mining is widely applied to the prediction of stock prices from financial time-series. However, few studies have examined the potential of data mining for shedding light on the qualitative problem-solving knowledge of experts who make stock price predictions. This paper presents a GA-based data mining approach to characterizing the qualitative knowledge of such experts, based on their observed predictions. This study is the first of its kind in the GA literature. The results indicate that this approach generates rules with higher accuracy and greater coverage than inductive learning methods or neural networks. They also indicate considerable agreement between the GA method and expert problem-solving approaches. Therefore, the proposed method offers a suitable tool for eliciting and representing expert decision rules, and thus constitutes an effective means of predicting the stock price index.

• The Journal of Asian Finance, Economics and Business
• /
• v.8 no.1
• /
• pp.7-14
• /
• 2021

This paper examines the impact of foreign ownership on stock price volatility in an emerging market, namely, Thailand. The data were obtained from SETSMART, the database of the Stock Exchange of Thailand (SET). After removing financial firms, banks, and insurance companies as well as filtering outliers, the final sample covers 1,755 firm-year observations from 371 nonfinancial firms listed on the SET over the five-year period from 2014 to 2018. The regression model consists of stock price volatility, measured by two methods, as the dependent variable, foreign ownership as the main independent variable, and firm characteristics including firm size, leverage, market-to book ratio, and stock turnover as the control variables. The pooled OLS, fixed effects, and random effects estimations are employed to examine the relationship between foreign ownership and stock price volatility. The results reveal that foreign ownership has a negative and significant impact on stock price volatility. The two-stage least squares (2SLS) are also performed to address potential endogeneity problem. The results still indicate a negative relationship between foreign ownership and stock price volatility. Taken together, the findings of this study suggest that foreign investors help reduce stock price volatility and thus stabilize share price in the Thai stock market.

• Survey Research
• /
• v.4 no.1
• /
• pp.49-71
• /
• 2003

The main purpose of this study was to propose new approaches for brand valuation using internet data. Data, which are necessary to evaluate the brand power, were consisted of the brand stock price data and the surveyed data through the internet. In this paper we examined the applicability of the internet data using correlation analysis between the brand stock price and the real stock price. We proposed new methods for relative evaluation of the brand power combining the brand stock price data with the surveyed data.

• The Journal of Industrial Distribution & Business
• /
• v.10 no.6
• /
• pp.51-58
• /
• 2019

Purpose - Prior studies reported that the opacity of information caused stock price crash. If managers fail to disclose unfavorable information about the firm over a long period of time, the stock price is overvalued compared to its original value. If the accumulated information reaches a critical point and spreads quickly to the market, the stock price plunges. Information management by management's disclosure policy can cause information uncertainty, which will lead to a plunge in stock prices in the future. Thus, this study aims at examining the impact of disclosure quality on crash risk by focusing on the unfaithful disclosure firms. Research design, data, and methodology - This study covers firms listed on KOSPI and KOSDAQ from 2004 to 2013. Firms excluded from the sample are non-December firms, capital-eroding firms, and financial firms. The financial data used in the research was extracted from the KIS-Value and TS2000 database. Unfaithful disclosure firm designation data was collected from the Korea Exchange's electronic disclosure system (kind.krx.co.kr). Stock crash is measured as a dummy variable that equals one if a firm experiences at least one crash week over the fiscal year, and zero otherwise. Results - Empirical results as to the relation between unfaithful disclosure corporation designation and stock price crashes are as follows: There was a significant positive association between unfaithful disclosure corporation designation and stock price crash. This result supports the hypothesis that firms that have previously exhibited unfaithful disclosure behavior are more likely to suffer stock price plunges due to information asymmetry. Second, stock price crashes due to unfaithful disclosures are more likely to occur in Chaebol firms. Conclusions - While previous studies used estimates as a proxy for information opacity, this study used an objective measure such as unfaithful disclosure corporation designation. The designation by Korea Exchange is an objective evidence that the firm attempted to conceal and distort information in the previous year. The results of this study suggest that capital market investors need to investigate firms' disclosure behaviors.

• Journal of Korea Multimedia Society
• /
• v.14 no.9
• /
• pp.1152-1164
• /
• 2011

Stock price is stream data to change continuously. The characteristics of these data, stock trends according to flow of time intervals may differ. therefore, stock price should be continuously prediction when the price is updated. In this paper, we propose the new prediction system that continuously predicts the stock price according to the predefined time intervals for the selected stock item using HTM model. We first present a preprocessor which normalizes the stock data and passes its result to the stream sensor. We next present a stream sensor which efficiently processes the continuous input. In addition, we devise a storage node which stores the prediction results for each level and passes it to next upper level and present the HTM network for prediction using these nodes. We show experimented our system using the actual stock price and shows its performance.