Action-specialized expert ensemble trading system with extended discrete action space using deep reinforcement learning.

Despite active research on trading systems based on reinforcement learning, the development and performance of research methods require improvements. This study proposes a new action-specialized expert ensemble method consisting of action-specialized expert models designed specifically for each reinforcement learning action: buy, hold, and sell. Models are constructed by examining and defining different reward values that correlate with each action under specific conditions, and investment behavior is reflected with each expert model. To verify the performance of this technique, profits of the proposed system are compared to those of single trading and common ensemble systems. To verify robustness and account for the extension of discrete action space, we compared and analyzed changes in profits of the three actions to our model's results. Furthermore, we checked for sensitivity with three different reward functions: profit, Sharpe ratio, and Sortino ratio. All experiments were conducted with S&P500, Hang Seng Index, and Eurostoxx50 data. The model was 39.1% and 21.6% more efficient than single and common ensemble models, respectively. Considering the extended discrete action space, the 3-action space was extended to 11- and 21-action spaces, and the cumulative returns increased by 427.2% and 856.7%, respectively. Results on reward functions indicated that our models are well trained; results of the Sharpe and Sortino ratios were better than the implementation of profit only, as in the single-model cases. The Sortino ratio was slightly better than the Sharpe ratio.

Estimating the Fiscal Costs of Osteoporosis in Korea Applying a Public Economic Perspective.

BACKGROUND: Osteoporosis and attributable fractures are disruptive health events that can cause short and long-term cost consequences for families, health service and government. In this fracture-based scenario analysis we evaluate the broader public economic consequences for the Korean government based on fractures that can occur at 3 different ages. METHODS: We developed a public economic modelling framework based on population averages in Korea for earnings, direct taxes, indirect taxes, disability payments, retirement, pension payments, and osteoporosis health costs. Applying a scenario analysis, we estimated the cumulative average per person fiscal consequences of osteoporotic fractures occurring at different ages 55, 65, and 75 compared to average non-fracture individuals of comparable ages to estimate resulting costs for government in relation to lost tax revenue, disability payments, pension costs, and healthcare costs. All costs are calculated between the ages of 50 to 80 in Korean Won (KRW) and discounted at 0.5%. RESULTS: From the scenarios explored, fractures occurring at age 55 are most costly for government with increased disability and pension payments of KRW 26,048,400 and KRW 41,094,206 per person, respectively, compared to the non-fracture population. A fracture can result in reduction in lifetime direct and indirect taxes resulting in KRW 53,648,886 lost tax revenue per person for government compared to general population. CONCLUSIONS: The fiscal consequences of osteoporotic fractures for government vary depending on the age at which they occur. Fiscal benefits for government are greater when fractures are prevented early due to the potential to prevent early retirement and keeping people in the labor force to the degree that is observed in non-fracture population.

Medical Treatment with Somatostatin Analogues in Acromegaly: Position Statement.

The Korean Endocrine Society (KES) published clinical practice guidelines for the treatment of acromegaly in 2011. Since then, the number of acromegaly cases, publications on studies addressing medical treatment of acromegaly, and demands for improvements in insurance coverage have been dramatically increasing. In 2017, the KES Committee of Health Insurance decided to publish a position statement regarding the use of somatostatin analogues in acromegaly. Accordingly, consensus opinions for the position statement were collected after intensive review of the relevant literature and discussions among experts affiliated with the KES, and the Korean Neuroendocrine Study Group. This position statement includes the characteristics, indications, dose, interval (including extended dose interval in case of lanreotide autogel), switching and preoperative use of somatostatin analogues in medical treatment of acromegaly. The recommended approach is based on the expert opinions in case of insufficient clinical evidence, and where discrepancies among the expert opinions were found, the experts voted to determine the recommended approach.

Forecasting stock prices with a feature fusion LSTM-CNN model using different representations of the same data.

Forecasting stock prices plays an important role in setting a trading strategy or determining the appropriate timing for buying or selling a stock. We propose a model, called the feature fusion long short-term memory-convolutional neural network (LSTM-CNN) model, that combines features learned from different representations of the same data, namely, stock time series and stock chart images, to predict stock prices. The proposed model is composed of LSTM and a CNN, which are utilized for extracting temporal features and image features. We measure the performance of the proposed model relative to those of single models (CNN and LSTM) using SPDR S&P 500 ETF data. Our feature fusion LSTM-CNN model outperforms the single models in predicting stock prices. In addition, we discover that a candlestick chart is the most appropriate stock chart image to use to forecast stock prices. Thus, this study shows that prediction error can be efficiently reduced by using a combination of temporal and image features from the same data rather than using these features separately.

Economic Burden of Osteoporosis in South Korea: Claim Data of the National Health Insurance Service from 2008 to 2011.

The purpose of this study was to estimate the current economic burden of osteoporosis in South Korea using national claim data of the Korean National Health Insurance Service (KNHIS) from 2008 to 2011. Patients aged 50 years or older were identified from KNHIS nationwide database for all records of outpatient visits or hospital admissions. Healthcare costs for osteoporotic patients included direct medical costs for hospitalization, outpatient care, and prescription drugs for the year after discharge. Healthcare costs were estimated based on the perspective of KNHIS, and calculated using a bottom-up approach. Between 2008 and 2011, total healthcare costs for osteoporotic patients increased from 3976 million USD to 5126 million USD, with an annual increase of 9.2% which accounted for one-sixth (16.7%) of national healthcare expenditure. Healthcare cost for hospitalization was the highest ($1903 million, 40.0% of total healthcare cost), followed by cost for outpatient care ($1474 million, 31.0%) and cost for prescription drugs ($1379 million, 29.0%). Although total healthcare cost for osteoporotic men was 6 times lower than that for osteoporotic women, the cost per person was 1.5 times higher than that for women. Total healthcare cost for osteoporotic patients without fractures was higher than that for osteoporotic patients with fractures. However, cost per person was the opposite. Osteoporosis entails substantial epidemiologic and economic burden in South Korea. This study provides information about the total healthcare burden, which could be important when determining what attention and awareness osteoporosis should be given in the public health system.

Healthcare Costs of Osteoporotic Fracture in Korea: Information from the National Health Insurance Claims Database, 2008-2011.

BACKGROUND: The present study estimated healthcare costs of osteoporotic fractures including spine, hip, distal radius and humerus in Koreans over 50 years of age using national claims data. METHODS: Korea National Health Insurance data between 2008 and 2011 was searched for all claims records of outpatient visits or hospital admissions of patients ≥50-year-of-age. Osteoporosis-related fractures were identified using certain the International Classification of Diseases, Tenth Revision codes and site-specific physician claims for procedures in a patient age cut-off value of 50 years. The healthcare costs included acute phase costs accounting for emergency medical care given immediately after fracture, costs due to further hospitalization and surgical procedures, physiotherapy sessions according to the site of the fracture, and outpatient visits in the year after discharge. RESULTS: The total estimated healthcare costs of osteoporotic fractures in 2011 was $722 million. From 2008 to 2011, the total number and healthcare costs of osteoporotic fractures increased 28.9% (from 127,070 to 163,823) and 31.6% (from $549 million to $722 million), respectively. The portion of national health care expenditure was ranged from 2.3% in 2008 to 2.2% in 2011. The mean healthcare cost of osteoporotic fractures per person increased 2.1% from $4,321 in 2008 to $4,410 in 2011.The mean healthcare costs were highest for hip fractures followed by spine, humerus, and distal radius fractures. CONCLUSIONS: Total Healthcare costs of osteoporotic fractures in South Koreans ≥50-year-of-age increased between 2008 and 2011. This trend will likely continue, which is an important health problem in the elderly population and economically.

Development of a Korean Fracture Risk Score (KFRS) for Predicting Osteoporotic Fracture Risk: Analysis of Data from the Korean National Health Insurance Service.

BACKGROUND: Asian-specific prediction models for estimating individual risk of osteoporotic fractures are rare. We developed a Korean fracture risk prediction model using clinical risk factors and assessed validity of the final model. METHODS: A total of 718,306 Korean men and women aged 50-90 years were followed for 7 years in a national system-based cohort study. In total, 50% of the subjects were assigned randomly to the development dataset and 50% were assigned to the validation dataset. Clinical risk factors for osteoporotic fracture were assessed at the biennial health check. Data on osteoporotic fractures during the follow-up period were identified by ICD-10 codes and the nationwide database of the National Health Insurance Service (NHIS). RESULTS: During the follow-up period, 19,840 osteoporotic fractures were reported (4,889 in men and 14,951 in women) in the development dataset. The assessment tool called the Korean Fracture Risk Score (KFRS) is comprised of a set of nine variables, including age, body mass index, recent fragility fracture, current smoking, high alcohol intake, lack of regular exercise, recent use of oral glucocorticoid, rheumatoid arthritis, and other causes of secondary osteoporosis. The KFRS predicted osteoporotic fractures over the 7 years. This score was validated using an independent dataset. A close relationship with overall fracture rate was observed when we compared the mean predicted scores after applying the KFRS with the observed risks after 7 years within each 10th of predicted risk. CONCLUSION: We developed a Korean specific prediction model for osteoporotic fractures. The KFRS was able to predict risk of fracture in the primary population without bone mineral density testing and is therefore suitable for use in both clinical setting and self-assessment. The website is available at http://www.nhis.or.kr.

Trends of Incidence, Mortality, and Future Projection of Spinal Fractures in Korea Using Nationwide Claims Data.

Spinal fractures have been recognized as a major health concern. Our purposes were to evaluate the trends in the incidence and mortality of spinal fractures between 2008 and 2012 and predict the number of spinal fractures that will occur in Korea up to 2025, using nationwide data from the National Health Insurance Service (NHIS). A nationwide data set was evaluated to identify all new visits to medical institutes for spinal fractures in men and women aged 50 years or older between 2008 and 2012. The incidence, mortality rates and estimates of the number of spinal fractures were calculated using Poisson regression. The number of spinal fractures increased over the time span studied. Men and women experienced 14,808 and 55,164 vertebral fractures in 2008 and 22,739 and 79,903 in 2012, respectively. This reflects an increase in the incidence of spinal fractures for both genders (men, 245.3/100,000 in 2008 and 312.5/100,000 in 2012; women, 780.6/100,000 in 2008 and 953.4/100,000 in 2012). The cumulative mortality rate in the first year after spinal fractures decreased from 8.51% (5,955/69,972) in 2008 to 7.0% (7,187/102,642) in 2012. The overall standardized mortality ratio (SMR) of spinal fractures at 1 year post-fracture was higher in men (7.76, 95% CI: 7.63-7.89) than in women (4.70, 95% CI: 4.63-4.76). The total number of spinal fractures is expected to reach 157,706 in 2025. The incidence of spinal fractures increased in Korea in the last 5 years, and the socioeconomic burden of spinal fractures will continue to increase in the near future.