• Journal of the Korea Society of Computer and Information
• /
• v.27 no.11
• /
• pp.147-155
• /
• 2022

In this study, we developed a system to dynamically balance a daily stock portfolio and performed trading simulations using gradient boosting and genetic algorithms. We collected various stock market data from stocks listed on the KOSPI and KOSDAQ markets, including investor-specific transaction data. Subsequently, we indexed the data as a preprocessing step, and used feature engineering to modify and generate variables for training. First, we experimentally compared the performance of three popular gradient boosting algorithms in terms of accuracy, precision, recall, and F1-score, including XGBoost, LightGBM, and CatBoost. Based on the results, in a second experiment, we used a LightGBM model trained on the collected data along with genetic algorithms to predict and select stocks with a high daily probability of profit. We also conducted simulations of trading during the period of the testing data to analyze the performance of the proposed approach compared with the KOSPI and KOSDAQ indices in terms of the CAGR (Compound Annual Growth Rate), MDD (Maximum Draw Down), Sharpe ratio, and volatility. The results showed that the proposed strategies outperformed those employed by the Korean stock market in terms of all performance metrics. Moreover, our proposed LightGBM model with a genetic algorithm exhibited competitive performance in predicting stock price movements.

• Journal of the Korean Data and Information Science Society
• /
• v.22 no.2
• /
• pp.255-267
• /
• 2011

The aim of this study is to utilize system trading for making investment decisions and use technical analysis and Dynamic Time Warping (DTW) to determine similar patterns in the frequency of stock data and ascertain the optimal timing for trade. The study will examine some of the most common patterns in the futures market and use DTW in terms of their frequency (10, 30, 60 minutes, and daily) to discover similar patterns. The recognized similar patterns were verified by executing trade simulation after applying specific strategies to the technical indicators. The most profitable strategies among the set of strategies applied to common patterns were again applied to the similar patterns and the results from DTW pattern recognition were examined. The outcome produced useful information on determining the optimal timing for trade by using DTW pattern recognition through system trading, and by applying distinct strategies depending on data frequency.

• Journal of the Korean Data and Information Science Society
• /
• v.24 no.1
• /
• pp.135-150
• /
• 2013

This study proposes a novel option strategy by using characteristic of volatility reversion and rough set algorithm in options market. Until now, various research has been conducted on stock and future markets, but minimal research has been done in options market. Particularly, research on the option trading strategy using high frequency data is limited. This study consists of two purposes. The first is to enjoy a profit using volatility reversion model when volatility gap is occurred. The second is to pursue a more stable profit by filtering inaccurate entry point through rough set algorithm. Since options market is affected by various elements like underlying assets, volatility and interest rate, the point of this study is to hedge elements except volatility and enjoy the profit following the volatility gap.

• Journal of Intelligence and Information Systems
• /
• v.20 no.4
• /
• pp.43-58
• /
• 2014

As the use of trading systems has increased rapidly, many researchers have become interested in developing effective stock market prediction models using artificial intelligence techniques. Stock market prediction involves multifaceted interactions between market-controlling factors and unknown random processes. A successful stock prediction model achieves the most accurate result from minimum input data with the least complex model. In this research, we develop a combination model of ${\pi}$-fuzzy logic and support vector machine (SVM) models, using a genetic algorithm to optimize the parameters of the SVM and ${\pi}$-fuzzy functions, as well as feature subset selection to improve the performance of stock market prediction. To evaluate the performance of our proposed model, we compare the performance of our model to other comparative models, including the logistic regression, multiple discriminant analysis, classification and regression tree, artificial neural network, SVM, and fuzzy SVM models, with the same data. The results show that our model outperforms all other comparative models in prediction accuracy as well as return on investment.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.36 no.4
• /
• pp.123-129
• /
• 2013

The aim of this study is to design an intelligent pattern-based real-time trading system (PRTS) using rough set analysis of technical indicators, dynamic time warping (DTW), and genetic algorithm in stock futures market. Rough set is well known as a data-mining tool for extracting trading rules from huge data sets such as real-time data sets, and a technical indicator is used for the construction of the data sets. To measure similarity of patterns, DTW is used over a given period. Through an empirical study, we identify the ideal performances that were profitable in various market conditions.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.42 no.3
• /
• pp.13-24
• /
• 2017

This paper aims to implement "structural changes detection procedure" in pairs trading algorithm and to show that the proposed approach outperforms the extant pair trading algorithm. Structural changes in pairs trading are defined in terms of changes in cointegrating factors and broken cointegration relationship. These changes are designed to test extant structural changes and unit root test methodologies. The simulation finds that expanding the changes in structure, increasing the mean reverting process of spread, and extending the consecutive days of broken cointegration will increase the performances of the proposed algorithm. Empirical study results are also consistent those of the simulation studies. The proposed algorithm outperforms the extant algorithm relative to risk and return given that the cumulative profit/loss has a significant upward-slope with minimal variance.

• The KIPS Transactions:PartA
• /
• v.9A no.2
• /
• pp.241-248
• /
• 2002

With the structure of tilde-area computing system which Is specified by a researching team in Vrije University, Netherlands, lots of researchers and developers have been progressing the studies of global location and interconnection services of distributed objects existing in global sites. Most of them halve focused on binding services of only non-duplicated computational objects existing wide-area computing sites without any consideration of duplication problems. But all of objects existing on the earth rave the duplicated characteristics according to how to categorize their own names or properties. These objects with the same property can define as duplicated computational objects. Up to now, the existing naming or trading mechanism has not supported the binding services of duplicated objects, because of deficiency of independent location service. For this reason, we suggest a new model that can not only manages locations of duplicated objects In wide-area computing environments, but also provide minimum binding time by considering both the optimal selection of one of duplicated objects and load balance among distributed systems. Our model is functionally divided into 2 parts, one part to obtain an unique object handle of duplicated objects with same property as a naming and trading service, and the other to search one or more contact addresses by a node manager using a liven object handle, as a location service For location transparency, these services are independently executing each other. Based on our model, we described structure of wide-area integrated tree and algorithms for searching and updating contact address of distributed object on this tree. finally, we showed a federation structure that can globally bind distributed objects located on different regions from an arbitrary client object.

• Journal of the Korea Convergence Society
• /
• v.10 no.9
• /
• pp.199-207
• /
• 2019

This study aims to analyze the profitability of Robo-Advisors portfolio combined with the analysts' forecasts on the Korean stock prices. Sample stocks are 8 blue-chips and sample period is from 2003 to 2019. Robo-Advisor portfolio was suggested using the Black-Litterman model combined with the analysts' forecasts and its profitability was analyzed. Empirical result showed the suggested Robo-Advisor algorithm produced 1% annual excess return more than that of the benchmark. The study documented that the analysts' forecasts had an economic value when applied in the Robo-Advisor portfolio despite the prevalent blames from investors. The profitability on small or medium-sized stocks will need to be analyzed in the Robo-Advisor context because their information is relatively less known to investors and as such is expected to be strongly influenced by the analysts' forecasts.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.1
• /
• pp.123-130
• /
• 2021

With the recent developments in computer technology, there has been an increasing interest in the field of machine learning. This also has led to a significant increase in real business cases of machine learning theory in various sectors. In finance, it has been a major challenge to predict the future value of financial products. Since the 1980s, the finance industry has relied on technical and fundamental analysis for this prediction. For future value prediction models using machine learning, model design is of paramount importance to respond to market variables. Therefore, this paper quantitatively predicts the stock price movements of individual stocks listed on the KOSPI market using machine learning techniques; specifically, the reinforcement learning model. The DQN and A2C algorithms proposed by Google Deep Mind in 2013 are used for the reinforcement learning and they are applied to the stock trading strategies. In addition, through experiments, an input value to increase the cumulative profit is selected and its superiority is verified by comparison with comparative algorithms.

• The Journal of Information Systems
• /
• v.26 no.3
• /
• pp.63-90
• /
• 2017

Purpose This study proposes a novel system trading model using case-based reasoning (CBR) based on absolute similarity threshold. The proposed model is designed to optimize the absolute similarity threshold, feature selection, and instance selection of CBR by using genetic algorithm (GA). With these mechanisms, it enables us to yield higher returns from stock market trading. Design/Methodology/Approach The proposed CBR model uses the absolute similarity threshold varying from 0 to 1, which serves as a criterion for selecting appropriate neighbors in the nearest neighbor (NN) algorithm. Since it determines the nearest neighbors on an absolute basis, it fails to select the appropriate neighbors from time to time. In system trading, it is interpreted as the signal of 'hold'. That is, the system trading model proposed in this study makes trading decisions such as 'buy' or 'sell' only if the model produces a clear signal for stock market prediction. Also, in order to improve the prediction accuracy and the rate of return, the proposed model adopts optimal feature selection and instance selection, which are known to be very effective in enhancing the performance of CBR. To validate the usefulness of the proposed model, we applied it to the index trading of KOSPI200 from 2009 to 2016. Findings Experimental results showed that the proposed model with optimal feature or instance selection could yield higher returns compared to the benchmark as well as the various comparison models (including logistic regression, multiple discriminant analysis, artificial neural network, support vector machine, and traditional CBR). In particular, the proposed model with optimal instance selection showed the best rate of return among all the models. This implies that the application of CBR with the absolute similarity threshold as well as the optimal instance selection may be effective in system trading from the perspective of returns.