• Journal of Intelligence and Information Systems
• /
• v.25 no.3
• /
• pp.239-251
• /
• 2019

Statistical methods such as moving averages, Kalman filtering, exponential smoothing, regression analysis, and ARIMA (autoregressive integrated moving average) have been used for stock market predictions. However, these statistical methods have not produced superior performances. In recent years, machine learning techniques have been widely used in stock market predictions, including artificial neural network, SVM, and genetic algorithm. In particular, a case-based reasoning method, known as k-nearest neighbor is also widely used for stock price prediction. Case based reasoning retrieves several similar cases from previous cases when a new problem occurs, and combines the class labels of similar cases to create a classification for the new problem. However, case based reasoning has some problems. First, case based reasoning has a tendency to search for a fixed number of neighbors in the observation space and always selects the same number of neighbors rather than the best similar neighbors for the target case. So, case based reasoning may have to take into account more cases even when there are fewer cases applicable depending on the subject. Second, case based reasoning may select neighbors that are far away from the target case. Thus, case based reasoning does not guarantee an optimal pseudo-neighborhood for various target cases, and the predictability can be degraded due to a deviation from the desired similar neighbor. This paper examines how the size of learning data affects stock price predictability through k-nearest neighbor and compares the predictability of k-nearest neighbor with the random walk model according to the size of the learning data and the number of neighbors. In this study, Samsung electronics stock prices were predicted by dividing the learning dataset into two types. For the prediction of next day's closing price, we used four variables: opening value, daily high, daily low, and daily close. In the first experiment, data from January 1, 2000 to December 31, 2017 were used for the learning process. In the second experiment, data from January 1, 2015 to December 31, 2017 were used for the learning process. The test data is from January 1, 2018 to August 31, 2018 for both experiments. We compared the performance of k-NN with the random walk model using the two learning dataset. The mean absolute percentage error (MAPE) was 1.3497 for the random walk model and 1.3570 for the k-NN for the first experiment when the learning data was small. However, the mean absolute percentage error (MAPE) for the random walk model was 1.3497 and the k-NN was 1.2928 for the second experiment when the learning data was large. These results show that the prediction power when more learning data are used is higher than when less learning data are used. Also, this paper shows that k-NN generally produces a better predictive power than random walk model for larger learning datasets and does not when the learning dataset is relatively small. Future studies need to consider macroeconomic variables related to stock price forecasting including opening price, low price, high price, and closing price. Also, to produce better results, it is recommended that the k-nearest neighbor needs to find nearest neighbors using the second step filtering method considering fundamental economic variables as well as a sufficient amount of learning data.

• Proceedings of the KSR Conference
• /
• 2011.05a
• /
• pp.1352-1357
• /
• 2011

Heights of a platform above the rail for the passenger train in the country are classified into two categories such as the low level (500mm; mainline) and the high level (1,135mm; metropolitan subway line) platforms. In order to operate similarly both a mainline railroad and a metropolitan subway line, as the requisite door safety system, it is necessary to develop the doorstep equipment of the rolling stock regardless of both the low and high level platforms. In this study, the application of doorstep equipments to use mixed with two types of platforms are examined on the supposition that the train only for the low level platform stops in the both low and high level platforms.

• Proceedings of the Korea Concrete Institute Conference
• /
• 2004.05a
• /
• pp.168-171
• /
• 2004

If a research trend present in and outside the country is often seen, the structure measurement method of having used PZT and the optical fiber (FBG) will be the actual condition which has accomplished the stock. In order to manage such cracks, time, efforts and expense are required. Such a method has many difficulties in application of a structure by the difficult problem of the measurement range, and the expensive sensor price. Progressive cracks were generated by fracture of glass pipe sensor. Moreover, the experiment which can detect damage propriety by external Light Emitting Diode by damaging a glass pipe by load change with the application of switch using strain gage of a glass pipe was conducted.

• Journal of Korea Technical Association of The Pulp and Paper Industry
• /
• v.35 no.3
• /
• pp.29-36
• /
• 2003

Reducing the energy consumption while maintaining pulp quality is an important objective of today's paper industry. Enzymatic treatment of fibers and the application of dry strength agent were investigated as methods to reduce energy consumption during refining and to upgrade fiber characteristics. Modification of recycled fibers with an enzyme was effective in improving relining efficiency and reducing refining energy. Optimization of dry strength agent application conditions including stock pH, cationic demand, zeta potential, etc. were found to be very important for improving its effectiveness.

• The Korean Journal of Applied Statistics
• /
• v.23 no.2
• /
• pp.249-261
• /
• 2010

Long-term forecasting of seasonal time series is critical in many applications such as planning business strategies and resolving possible problems of a business company. Unlike the traditional approach that depends solely on dynamic models, Li and Hinich (2002) introduced a combination of stochastic dynamic modeling with filter bank approach for forecasting seasonal patterns using highly coherent(High-C) waveforms. We modify the filter selection and forecasting procedure on wavelet domain to be more feasible and compare the resulting predictor with one that obtained from the wavelet variance estimation method. An improvement over other seasonal pattern extraction and forecasting methods based on such as wavelet scalogram, Holt-Winters, and seasonal autoregressive integrated moving average(SARIMA) is shown in terms of the prediction error. The performance of the proposed method is illustrated by a simulation study and an application to the real stock price data.

• Wind and Structures
• /
• v.13 no.3
• /
• pp.221-234
• /
• 2010

In light-frame wood construction, missing roof-sheathing fasteners can be a relatively common occurrence. This type of construction makes up the vast majority of the residential building stock in North America and thus their performance in high winds, including hurricanes, is of concern due to their sheer number. Construction quality issues are common in these types of structures primarily because the majority are conventionally constructed and unlike steel and reinforced concrete structures, inspection is minimal except in certain areas of the country. The concept of performance-based wind engineering (PBWE), a relatively new paradigm, relies on the assumption that building performance under wind loads can be accurately modeled. However, the discrepancy between what is designed (and modeled) and what is built (the as-built) may make application of PBWE to light-frame wood buildings quite difficult. It can be concluded from this study that construction quality must be controlled for realistic application of PBWE to light-frame wood buildings.

• Korean Journal of Organic Agriculture
• /
• v.18 no.3
• /
• pp.419-432
• /
• 2010

This study was conducted to determine effects of applying cattle slurry and mixed sowing with legumes such as hairy vetch or forage pea on productivity, feed values and organic stock carrying capacity of whole crop barley and rye as winter forage crops, and to obtain organic forages together with higher soil fertility. Experimental plots consisted of 7 treatments, which were non-fertilizer, chemical fertilizer (containing phosphate and potassium: P+K), chemical fertilizer (containing nitrogen, phosphate and potassium: N+P+K), organic fertilizer, cattle slurry, cattle slurry application (mixture with hairy vetch), and cattle slurry application (mixture with forage pea) plots. Each treatment was triplicates, and seven treatments were allocated in a completely randomized block design. For whole crop barley or its mixture crops, annual dry matter (DM), crude protein (CP), and total digestible nutrients (TDN) yields of N+P+K plots were significantly (P<0.05) higher than other plots except for cattle slurry plots. The CP content of barley or its mixture crops was significantly higher tor N+P+K plot (9.8%) and mixture plots with legumes (8.6~9.7%) than those of other treatments. As 450 kg Hanwoo heifers were fed diets included 70% whole crop barley or 70% mixture crops with legumes, mixture plots are capable of raising average 1.7 to 1.8 heads/ha a year. For rye or its mixture crops, annual DM, CP, and TDN yields represented 6.9~7.1, 0.5~0.6 and 4.3~4.4 ton/ha, respectively. The N+P+K plot contained 10.8% CP, which was higher (P<0.05) than all other treatments. In case of 450 kg Hanwoo heifers fed diets included 70% rye or 70% mixture crops with legumes, mixture plots can rear average 1.9 heads/ha a year. When it was considered based on crop yields and organic stock carrying capacity, applying cattle slurry to whole crop barley or rye had the comparable yields and feed values to chemical fertilizer application. Moreover, whole crop barley and rye within cattle slurry plots had a greater combination with hairy vetch and forage pea, respectively, and their mixture crops with legumes had higher crude protein and TDN yields within cattle slurry plots. In conclusion, it would be expected that mixed sowing with legumes in the application of cattle slurry to grass crops could be substituted for imported organic grains as dietary protein sources in feeding organic livestock.

• Journal of Korean Society of Forest Science
• /
• v.100 no.4
• /
• pp.591-597
• /
• 2011

Forest carbon stock changes in a national forest were assessed by CBM-CFS3 model with different management scenarios to support decision making for a long term forest planning. Management scenarios were composed with 4 different levels of timber harvesting - current harvesting level (scenario1), 30% increment in each period (scenario2), 3 times increment (scenario3), and 5 times increment (scenario4). For each scenarios, changes in total carbon stocks, carbon stocks of each carbon pools, carbon stocks of harvested wood products (HWP) and age class structure were estimated over 100-year planning horizon. The estimated total carbon stock including HWP at the end of final period (100 years) was 433.1 tC/ha under scenario 1, but the age class structure has skewed right to the upper classes, which is not desirable for sustainable forest management. Under the scenario 4, however, the total carbon stock decrease to 385.5 tC/ha and the area of old growth forest show a significant decline. The estimated total carbon stock under scenario 2 and 3 were 411.7 tC/ha and 410.5 tC/ha respectively, and it was able to maintain the initial level of the forest carbon stocks during the planning horizon. Also the age class structures under the scenario 2 and 3 were evenly distributed from class 1 to class 8. Overall, scenario 2 and 3 were the most acceptable forest management options, in terms of carbon stock changes and age class structure.

• Korean Journal of Organic Agriculture
• /
• v.21 no.2
• /
• pp.247-263
• /
• 2013

This study was conducted to investigate the effect of cattle manure application on productivity, feed value, and stock carrying capacity of forage crops in upland and paddy fields at Gang-Wondo area. In the result, dry matter yield of sorghum ${\times}$ sudangrass hybrids obtained was 15.12 ton/ha at the level of 150kg N/ha of composted cattle manure. Significantly highest values of crude protein and total digestible nutrients (TDN) yields obtained were 0.59 and 5.35 ton/ha at the level of 150kg N/ha of composted cattle manure in the first cutting, and 0.44 and 3.70 ton/ha at the level of 150kg N/ha of organic raw cattle manure in the second cutting, respectively. The values of $K_{CP}+K_{TDN}/2$ and $K_{ME}$ of sorghum ${\times}$ sudangrass hybrids obtained was 7.76 and 4.46 head/ha at the level of 150kg N/ha of composted cattle manure. The dry matter yield, crude protein and TDN yields of rice straw were 4.95, 0.16 and 2.75 ton/ha at the level of 100kg N/ha of organic raw cattle manure, and the values of $K_{CP}+K_{TDN}/2$ and $K_{ME}$ of rice straw were 1.89 and 3.43 head/ha. The dry matter yield of winter crops, rye+red clover was 4.36 ton/ha in upland field, and rye+hairy vetch was 4.19 ton/ha in paddy field at the level of 100kg N/ha of composted cattle manure. Crude protein and TDN yields of rye+red clover was 0.29 and 2.38 ton/ha at the level of 100kg N/ha of composted cattle manure in upland field, and rye+hairy vetch was 0.30 and 2.48 ton/ha at the level of 80kg N/ha of composted cattle manure in paddy field. The values of $K_{CP}+K_{TDN}/2$ and $K_{ME}$ of rye+red clover was 2.34 and 2.15 head/ha in upland field, and rye+hairy vetch were 2.27 and 2.11 head/ha in paddy field, respectively. As the result, the productivity, feed value, and stock carrying capacity of sorghum ${\times}$ sudangrass hybrids showed higher values with composted cattle manure than organic raw cattle manure. rye+red clover in upland field and rye+hairy vetch in paddy field were most adaptable mixed combinations for roughage production at Gang-wondo area, it may due to their highly productivity, feed value, and stock carrying capacity.

• Journal of Intelligence and Information Systems
• /
• v.25 no.1
• /
• pp.63-83
• /
• 2019

Investors prefer to look for trading points based on the graph shown in the chart rather than complex analysis, such as corporate intrinsic value analysis and technical auxiliary index analysis. However, the pattern analysis technique is difficult and computerized less than the needs of users. In recent years, there have been many cases of studying stock price patterns using various machine learning techniques including neural networks in the field of artificial intelligence(AI). In particular, the development of IT technology has made it easier to analyze a huge number of chart data to find patterns that can predict stock prices. Although short-term forecasting power of prices has increased in terms of performance so far, long-term forecasting power is limited and is used in short-term trading rather than long-term investment. Other studies have focused on mechanically and accurately identifying patterns that were not recognized by past technology, but it can be vulnerable in practical areas because it is a separate matter whether the patterns found are suitable for trading. When they find a meaningful pattern, they find a point that matches the pattern. They then measure their performance after n days, assuming that they have bought at that point in time. Since this approach is to calculate virtual revenues, there can be many disparities with reality. The existing research method tries to find a pattern with stock price prediction power, but this study proposes to define the patterns first and to trade when the pattern with high success probability appears. The M & W wave pattern published by Merrill(1980) is simple because we can distinguish it by five turning points. Despite the report that some patterns have price predictability, there were no performance reports used in the actual market. The simplicity of a pattern consisting of five turning points has the advantage of reducing the cost of increasing pattern recognition accuracy. In this study, 16 patterns of up conversion and 16 patterns of down conversion are reclassified into ten groups so that they can be easily implemented by the system. Only one pattern with high success rate per group is selected for trading. Patterns that had a high probability of success in the past are likely to succeed in the future. So we trade when such a pattern occurs. It is a real situation because it is measured assuming that both the buy and sell have been executed. We tested three ways to calculate the turning point. The first method, the minimum change rate zig-zag method, removes price movements below a certain percentage and calculates the vertex. In the second method, high-low line zig-zag, the high price that meets the n-day high price line is calculated at the peak price, and the low price that meets the n-day low price line is calculated at the valley price. In the third method, the swing wave method, the high price in the center higher than n high prices on the left and right is calculated as the peak price. If the central low price is lower than the n low price on the left and right, it is calculated as valley price. The swing wave method was superior to the other methods in the test results. It is interpreted that the transaction after checking the completion of the pattern is more effective than the transaction in the unfinished state of the pattern. Genetic algorithms(GA) were the most suitable solution, although it was virtually impossible to find patterns with high success rates because the number of cases was too large in this simulation. We also performed the simulation using the Walk-forward Analysis(WFA) method, which tests the test section and the application section separately. So we were able to respond appropriately to market changes. In this study, we optimize the stock portfolio because there is a risk of over-optimized if we implement the variable optimality for each individual stock. Therefore, we selected the number of constituent stocks as 20 to increase the effect of diversified investment while avoiding optimization. We tested the KOSPI market by dividing it into six categories. In the results, the portfolio of small cap stock was the most successful and the high vol stock portfolio was the second best. This shows that patterns need to have some price volatility in order for patterns to be shaped, but volatility is not the best.