• Journal of Intelligence and Information Systems
• /
• v.23 no.2
• /
• pp.107-122
• /
• 2017

Volatility in the stock market returns is a measure of investment risk. It plays a central role in portfolio optimization, asset pricing and risk management as well as most theoretical financial models. Engle(1982) presented a pioneering paper on the stock market volatility that explains the time-variant characteristics embedded in the stock market return volatility. His model, Autoregressive Conditional Heteroscedasticity (ARCH), was generalized by Bollerslev(1986) as GARCH models. Empirical studies have shown that GARCH models describes well the fat-tailed return distributions and volatility clustering phenomenon appearing in stock prices. The parameters of the GARCH models are generally estimated by the maximum likelihood estimation (MLE) based on the standard normal density. But, since 1987 Black Monday, the stock market prices have become very complex and shown a lot of noisy terms. Recent studies start to apply artificial intelligent approach in estimating the GARCH parameters as a substitute for the MLE. The paper presents SVR-based GARCH process and compares with MLE-based GARCH process to estimate the parameters of GARCH models which are known to well forecast stock market volatility. Kernel functions used in SVR estimation process are linear, polynomial and radial. We analyzed the suggested models with KOSPI 200 Index. This index is constituted by 200 blue chip stocks listed in the Korea Exchange. We sampled KOSPI 200 daily closing values from 2010 to 2015. Sample observations are 1487 days. We used 1187 days to train the suggested GARCH models and the remaining 300 days were used as testing data. First, symmetric and asymmetric GARCH models are estimated by MLE. We forecasted KOSPI 200 Index return volatility and the statistical metric MSE shows better results for the asymmetric GARCH models such as E-GARCH or GJR-GARCH. This is consistent with the documented non-normal return distribution characteristics with fat-tail and leptokurtosis. Compared with MLE estimation process, SVR-based GARCH models outperform the MLE methodology in KOSPI 200 Index return volatility forecasting. Polynomial kernel function shows exceptionally lower forecasting accuracy. We suggested Intelligent Volatility Trading System (IVTS) that utilizes the forecasted volatility results. IVTS entry rules are as follows. If forecasted tomorrow volatility will increase then buy volatility today. If forecasted tomorrow volatility will decrease then sell volatility today. If forecasted volatility direction does not change we hold the existing buy or sell positions. IVTS is assumed to buy and sell historical volatility values. This is somewhat unreal because we cannot trade historical volatility values themselves. But our simulation results are meaningful since the Korea Exchange introduced volatility futures contract that traders can trade since November 2014. The trading systems with SVR-based GARCH models show higher returns than MLE-based GARCH in the testing period. And trading profitable percentages of MLE-based GARCH IVTS models range from 47.5% to 50.0%, trading profitable percentages of SVR-based GARCH IVTS models range from 51.8% to 59.7%. MLE-based symmetric S-GARCH shows +150.2% return and SVR-based symmetric S-GARCH shows +526.4% return. MLE-based asymmetric E-GARCH shows -72% return and SVR-based asymmetric E-GARCH shows +245.6% return. MLE-based asymmetric GJR-GARCH shows -98.7% return and SVR-based asymmetric GJR-GARCH shows +126.3% return. Linear kernel function shows higher trading returns than radial kernel function. Best performance of SVR-based IVTS is +526.4% and that of MLE-based IVTS is +150.2%. SVR-based GARCH IVTS shows higher trading frequency. This study has some limitations. Our models are solely based on SVR. Other artificial intelligence models are needed to search for better performance. We do not consider costs incurred in the trading process including brokerage commissions and slippage costs. IVTS trading performance is unreal since we use historical volatility values as trading objects. The exact forecasting of stock market volatility is essential in the real trading as well as asset pricing models. Further studies on other machine learning-based GARCH models can give better information for the stock market investors.

• Tuberculosis and Respiratory Diseases
• /
• v.45 no.3
• /
• pp.536-545
• /
• 1998

Background: The studies on prediction equations of pulmonary function parameters for adults in Korea have been performed in a reference population mainly consisted of young and middle ages. So they included a relatively few elderly who conducted pulmonary function test frequently in clinic. We established prediction equations of pulmonary function parameters for healthy adults over 50 years old in rural area and compared this results with those of other studies. Therefore we attempted to consider normative values of pulmonary function tests for elderly in Korea. Method: Five hundred thirty-three women and men over 50 years old in rural area were participated. A "healthy" subgroup of 110 women and 32 men were identified by excluding those who had conditions that negatively influenced pulmonary function. We derived prediction equations for FVC, $FEV_1$ and $FEV_1%$ by multiple linear regression method from their age, heights and weights in each sex. Results: Prediction equations for FVC and $FEV_1$ in each sex were derived as follows Male; FVC (L)=0.02488Height(cm)-0.0269Age(years)+0.493 $FEV_1(L)$=0.01874Weight(kg)-0.0282Age(years)+2.906 Female; FVC(L)=0.02160Height(cm)-0.0192Age(years)-0.0125 $FEV_1(L)$=0.01720Height(cm)-0.0194Age(years)+0.3890 Prediction equations for $FEV_1%$ were not derived because $FEV_1%$ didn't have statistically significant terms. Comparing Predicted values that were calculated by substitution into the equations of various studies of mean values of age, heights and weights from this study, FVC and $FEV_1$ values in men of this study were lower than those of other studies. In women, FVC and $FEV_1$ values of this study were as similar as or lower than those of the study conducted for healthy elderly blacks in U.S.A respectively. Conclusion: We have got prediction equations of pulmonary function parameters which were driven from forced expiratory spirogram in adults over 50 years old in rural area. Predicted values of this study were lower than those of other studies which were conducted in Korea. So we consider that the study for spirometry reference values for elderly Korean using the method compatible with ATS recommendation need to be conducted more frequently forward.

• Journal of Environmental Science International
• /
• v.23 no.2
• /
• pp.291-302
• /
• 2014

We have studied characteristics of community structure on the benthic macro-invertebrates at twelve selected sites in upstream, midstream and downstream of Paemsagol, Piagol, Chilseongol and Daeseonggol, major streams of Jirisan Mountain, for three years from April 2008 to September 2010. As a result, 2 phyla, 3 classes, 11 orders, 41 families, 119 species and 4,449 individuals of benthic macro-invertebrates have been collected and classified in this study. Among them EPT group (Ephemeroptera, Plecoptera, Trichoptera), commonly appearing taxa in clean stream ecosystem, was found most frequently. The highest species number of benthic macro-invertebrates were collected in 2008, the year with lower climatic influence including heavy rain. In almost all of survey sites 80% or over of EPT group were counted, and St. 1 scored highest rate of EPT group at 92.59%. Most dominant species was Ecdyonurus kibumensis at 12.83% of dominance index, and subdominant species was Goerodes KUb at 6.81% of dominance index. The investigation of the feeding function group indicated that gathering-collectors were dominant with 26 species and 1,334 individuals, while the number of filtering-collectors was lowest with 11 species and 230 individuals. This result represents that the feeding function group of Jirisan Mountain is mountainous stream-specific one. Both diversity index and richness index were lowest at the midstream of Chilseongol (St. 5), while the midstreams of Daeseonggol (St. 11) and Piagol (St. 8) were highest in diversity and richness indices respectively. The analyses of linear regression and correlation were performed in order to investigate and to predict the appearance aspect of EPT group by altitude. The results showed that the ratio of Plecopteran species number has increased by the elevation of the altitude at 0.05 significance level. Cluster analysis was also carried out for evaluating environmental similarities among survey sites. As a result, upper regions of Paemsagol (St. 1) and Piagol have clustered as most similar sites each other, while the midstream of Piagol (St. 8) has separated with lower similarity value than other sites in species composition.

• Journal of Environmental Health Sciences
• /
• v.33 no.5
• /
• pp.365-371
• /
• 2007

To evaluate the effect of air pollution on respiratory health in children, We conducted a longitudinal study in which children were asked to record their daily levels of Peak Expiratory Flow Rate(PEFR) using potable peak flow meter(mini-Wright) for 4 weeks. The relationship between daily PEFR and ambient air particle levels was analyzed using a mixed linear regression models including gender, age in year, weight, the presence of respiratory symptoms, and relative humidity as an extraneous variable. The daily mean concentrations of $PM_{10}$ and $PM_{2.5}$ over the study period were $64.9{\mu}g/m^3$ and $46.1{\mu}g/m^3$, respectively. The range of daily measured PEFR in this study was $182{\sim}481\;l/min$. Daily mean PEFR was regressed with the 24-hour average $PM_{10}(or\;PM_{2.5})$ levels, weather information such as air temperature and relative humidity, and individual characteristics including sex, weight, and respiratory symptoms. The analysis showed that the increase of air particle concentrations was negatively associated with the variability in PEFR. We estimated that the IQR increment of $PM_{10}$ or $PM_{2.5}$ were associated with 1.5 l/min (95% Confidence intervals -3.1, 0.1) and 0.8 l/min(95% CI -1.8, 0.1) decline in PEFR. Even though this study showed negative findings on the relationship between respiratory function and air particles, it was worth noting that the findings must be interpreted cautiously because exposure measurement based on monitoring of ambient air likely resulted in misclassification of true exposure levels and this was the first Korean study that $PM_{2.5}$ measurement was applied as an index of air quality.

• Journal of Environmental Health Sciences
• /
• v.30 no.2
• /
• pp.140-148
• /
• 2004

To evaluate the acute effects of fine particles on pulmonary function, a longitudinal study was conducted. This study was carried out for the schoolchildren (3rd and 6th grades) living in Beijing, China. Each child was provided with a mini-Wright peak flow meter and a preformatted health symptom diary for 40 days, and was trained on their proper use. Participants were instructed to perform the peak flow test three times in standing position, three times a day (9 am, 12 pm, and 8 pm), and to record all the readings along with the symptoms (cold, cough, and asthmatic symptoms) experienced on that day. Daily measurement of fine particles (PM$_{10}$ and PM$_{2.5}$) was obtained in the comer of the playground of the participating elementary school for the same period of this longitudinal study. The relationship between daily peak expiratory flow rate (PEFR) and fine particle levels was analyzed using a mixed linear regression models including gender, height, the presence of respiratory symptoms, and daily average temperature and relative humidity as extraneous variables. The total number of students participating in this longitudinal study was 87. The range of daily measured PEFR was 253-501$\ell$/min. In general, the PEFR measured in the morning was lower than the PEFR measured in the evening (or afternoon) on the same day. The daily mean concentrations of PM$_{10}$ and PM$_{2.5}$ over the study period were 180.2$\mu\textrm{g}$/㎥ and 103.2$\mu\textrm{g}$/㎥, respectively. The IQR (inter-quartile range) of PM$_{10}$ and PM$_{2.5}$ were 91.8$\mu\textrm{g}$/㎥ and 58.0$\mu\textrm{g}$/㎥. During the study period, the national ambient air quality standard of 150$\mu\textrm{g}$/㎥ (for PM$_{10}$) was exceeded in 23 days (57.5%). The analysis showed that an increase of 1$\mu\textrm{g}$/㎥ of PM$_{10}$ corresponded to 0.59$\mu\textrm{g}$/㎥ increment of PM$_{2.5}$. Daily mean PEFR was regressed with the 24-hour average PM$_{10}$ (or PM$_{2.5}$) levels, weather information such as air temperature and relative humidity, and individual characteristics including gender, height, and respiratory symptoms. The analysis showed that the increase of fine particle concentrations was negatively associated with the variability in PEFR. The IQR increments of PM$_{10}$ or PM$_{2.5}$ (at 1-day time lag) were also shown to be related with 1.54 $\ell$/min (95% Confidence intervals: 0.94-2.14) and 1.56$\ell$/min (95% CI: 0.95-2.16) decline in PEFR.R.ine in PEFR.ine in PEFR.

• Dementia and Neurocognitive Disorders
• /
• v.21 no.1
• /
• pp.30-41
• /
• 2022

Background and Purpose: In this study we aimed to find the association between neuropsychological performance and body mass index (BMI) in patients with mild cognitive impairment (MCI). In addition, we investigated the effects of the apolipoprotein E (APOE) genotype in the relationship between the BMI and cognition in MCI. Methods: We enrolled a cohort of 3,038 subjects with MCI aged 65-90 from the Clinical Research Center for Dementia of South Korea and a dementia cohort of the Ewha Womans University Mokdong Hospital. MCI patients were classified into three subgroups according to the Asian standard of BMI. We compared cognitive performances between groups by one-way analysis of variance. To investigate the effects of the APOE genotype, we used multivariate linear regression models after adjusting for possible confounders. Results: Even though normal BMI groups were younger, had more females, and had less comorbidities, the higher BMI groups had better cognitive functions. Among subjects with APOE ε4 carriers, there was a positive relationship between the BMI and the memory task alone. Conclusions: Our findings suggested that higher BMI in patients with MCI were associated with better cognitive performance. The effects of the APOE ε4 genotype in the associations between BMI and cognition were distinguishing. Therefore, according to physical status, APOE ε4 genotype-specific strategies in the assessments and treatments may be necessary in elderly patients with MCI.

• Sleep Medicine and Psychophysiology
• /
• v.27 no.1
• /
• pp.16-23
• /
• 2020

Objectives: Sleep disturbance in the elderly is associated with cognitive decline. Sleep quality is known to deteriorate with age, and prospective studies seldom have examined the relationship between sleep quality and cognitive function. This study investigates the relationship between early sleep quality and cognitive function based on six-year follow-up data of community individuals older than 60 years. Methods: The participants included 622 community elderly people older than 60 years from Jinju-Si. The final analysis comprised 322 elderly people. Pittsburgh sleep quality index (PSQI) and the Korean version of Consortium to Establish a Registry for Alzheimer's Disease (CERAD-K) were used to assess early sleep quality and cognitive function after six years. Multiple linear regression analysis was performed to investigate the association between early sleep quality and cognitive function in the elderly. Results: Early sleep quality (PSQI) was significantly associated with the results of the digit span test, clock drawing test (clox 1), and word recall test after six years. Sleep quality (PSQI) decreased significantly after six years, and lower quality of sleep (PSQI) score was associated with higher digit span test score (β = -0.167, p = 0.026) and higher clock drawing test score (β = -0.157, p = 0.031). Lower quality of sleep (PSQI) score was associated with higher word recall test (β = -0.140, p = 0.039). Conclusion: The digit span test, word recall test, and clock drawing task (CLOX 1) shown to be significantly associated to sleep quality can be performed fast and easily in clinical practice. It is important to assess early cognitive function in the elderly with poor sleep quality, and further studies could suggest that these tests may be useful screening tests for early dementia in elderly with poor sleep quality.

• Tuberculosis and Respiratory Diseases
• /
• v.46 no.6
• /
• pp.786-794
• /
• 1999

Back ground : Arm span measurements provide a practical substitute for standing height to predict normal spirometric values in subjects unable to stand or those with a skeletal deformity such as kyphoscoliosis. The relationship between arm span and height has previously been reported as either a fixed ratio unaffected by age or as a regression equation in which the ratio varies as a function of age. The fixed ratio or regression equation is known to be specific for sex and race. Methods : We studied the relationship between standing height, arm span, and age in 381 Korean adult female subjects (ages 20 to 69 yrs) sampled in a general population. Results : The mean ratio for arm span to height is 1.004. Multiple linear analysis found arm span and age to be predictive of standing height (p=0.0001, $r^2$=0.76). We performed the analysis of the difference between the predicted height using either fixed ratio or regression equation and actual height. At the extremes of arm span and age, the ratio method either underestimated(at smaller arm span or younger age) or overestimated(at larger arm span or older age) as compared with actual height (p=0.0001). Conclusion : This results indicate that the estimated height using the fixed ratio method provides a less acceptable method of estimating height for the prediction of lung volumes in the Korean adult women when compared with the regression equations, especially at the extremes of stature or age.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.10 no.4
• /
• pp.158-166
• /
• 2008

The pumpkin fruit fly, Bactrocera depressa (Tephritidae: Diptera), is one of the most important pests in Cucurbitaceae plants. This study was conducted to investigate the basic ecology of B. depressa, and to develop a forecasting model for predicting the time of adult emergence in early season. In green pumpkin producing farms, the oviposition punctures caused by the oviposition of B. depressa occurred first between mid- and late July, peaked in late August, and then decreased in mid-September followed by disappearance of the symptoms in late September, during which oviposition activity of B. depressa is considered active. In full-ripened pumpkin producing farms, damaged fruits abruptly increased from early Auguest, because the decay of pumpkins caused by larval development began from that time. B. depressa produced a mean oviposition puncture of 2.2 per fruit and total 28.8-29.8 eggs per fruit. Adult emergence from overwintering pupae, which was monitored using a ground emergence trap, was first observed between mid- and late May, and peaked during late May to early June. The development times from overwintering pupae to adult emergence decreased with increasing temperature: 59.0 days at $15^{\circ}C$, 39.3 days at $20^{\circ}C$, 25.8 days at$25^{\circ}C$ and 21.4 days at $30^{\circ}C$. The pupae did not develop to adult at $35^{\circ}C$. The lower developmental threshold temperature was calculated as $6.8^{\circ}C$ by linear regression. The thermal constant was 482.3 degree-days. The non-linear model of Gaussian equation well explained the relationship between the development rate and temperature. The Weibull function provided a good fit for the distribution of development times of overwintering pupae. The predicted date of 50% adult emergence by a degree-day model showed one day deviation from the observed actual date. Also, the output estimated by rate summation model, which was consisted of the developmental model and the Weibull function, well pursued the actual pattern of cumulative frequency curve of B. depressa adult emergence. Consequently, it is expected that the present results could be used to establish the management strategy of B. depressa.

• Journal of The Korean Ophthalmological Society
• /
• v.58 no.3
• /
• pp.321-326
• /
• 2017

Purpose To identify the correspondence between the central sensitivity of several visual field (VF) tests and ganglion cell inner plexiform layer (GC-IPL) thickness in early glaucoma patients with parafoveal scotoma. Methods Fifty-seven eyes from 57 patients with glaucomatous optic neuropathy and parafoveal scotoma were analyzed using the standard automated perimetry (SAP) C10-2 test, the SAP C24-2 test, and the frequency doubling technology perimetry (FDT) C24-2 test. The correlation between the VF central sensitivity and the GC-IPL thickness from macular scans via optical coherence tomography was analyzed. Results The central sensitivity was $27.51{\pm}5.43dB$, $27.39{\pm}5.05dB$, and $22.09{\pm}5.08dB$ for SAP C24-2, SAP C10-2, and FDT C24-2, respectively. Mean GC-IPL thickness was $70.2{\pm}8.5{\mu}m$. Using regression analysis, the value of log $R^2$ between the logarithmic central sensitivity and GC-IPL thickness was 0.498, and the linear $R^2$ between the antilogarithmic central sensitivity and GC-IPL thickness in SAP C10-2 was 0.486, and both were statistically significant (p < 0.05). This relationship was stronger in early glaucoma patients compared to late glaucoma patients using SAP C10-2. Conclusions The structure-function relationship between GC-IPL thickness and central sensitivity was better with SAP C10-2, especially in early glaucoma patients, compared to other VF modalities.