• Journal of Korean Society of Forest Science
• /
• v.87 no.1
• /
• pp.40-49
• /
• 1998

This study aimed to provide a strategy for selecting an adequate combination of growth intervals(i.e. times between age $T_1$ and age $T_2$) to be used to improve the reality of the growth equation through obtaining better precision of parameter estimates. Variety of growth functions were fitted to the data and one equation which best fitted the data was chosen for the analysis. A modified Schumacher projection equation, selected as a best equation, that included dummy variables representing locality as a predictor variable was fitted for basal area and height equations with nonoverlapping growth interval and all possible growth interval data sets of Douglas-fir(Pseudotsuga menziesii Mirb.Franco). The data were measured in all parts of the South Island of New Zealand. It was found that the precision of parameter estimates was increased in both basal area and height equations by using data set which contained a range of measurement intervals from short to long term.

• The Journal of the Korea Contents Association
• /
• v.14 no.11
• /
• pp.102-110
• /
• 2014

It is a time to change the investment system for performances of co-producing according to development of the environment of performing arts in Korea. If we keep producing performances through the conventional method of the investment, we soon fall behind because production costs will increase by extension of the performance period. Investment companies generally require to the production company the investment principal guarantee as a condition for a investment. Producing companies have been producing performance with only ticket revenue and money of investment companies without their own money. Those two unreasonable things have raised a risk. So first of all, it is necessary to account marginal cost instead of average cost to decide more exactly open-running of performance. Second, it is necessary to change total cost as an investment parameter to production cost to avoid the unreasonable demands, such as a principal guarantee to production company. Therefore, we have to accept the budget planning of the United States to account the marginal cost and the production cost.

• Environmental and Resource Economics Review
• /
• v.21 no.4
• /
• pp.947-980
• /
• 2012

This study analyzes the long-run impacts of climate change on Korean agriculture and economy. We estimate the impacts of climate change on the productivities of major agricultural products including rice, dairy and livestock using both a simulation approach and a semiparametric econometric model. The former predicts a decline in productivity while the latter predicts an increase in productivity due to climate change, especially for rice. A recursive dynamic CGE model is used to analyze the general equilibrium impacts of productivity change under the two different scenarios, derived from the two productivity analysis approaches. The loss of GDP in 2050 is 0.2% or 0.02% of total GDP depending on the scenario. It is shown that the losses in dairy and livestock sectors are larger than that in rice sector, although the losses in those two non-rice sectors have been ignored by most existing works.

• The Korean Journal of Applied Statistics
• /
• v.20 no.3
• /
• pp.449-458
• /
• 2007

Bipolar disorder, also known as manic-depressive illness, is a brain disorder that causes unusual shifts in person's mood, energy, and ability to function. Compared with manic episode, the depression episode causes more serious results such as restless, loss of interest or pleasure, or thoughts of death or suicide and the cure rate of depression episode is lower than that of manic episode. Furthermore, a long term use of antidepressants in bipolar patients may result in manic episode. Our interest is to investigate the effect of antidepressant on switch of moods of bipolar patients and to estimate the transition probabilities of switch between moods, depression and (hypo) manic. In this study, three approaches are applied in terms of multi state model. Parametric model is applied using left censoring data and nonparametric model is implemented under illness-death model with counting process. In order to estimate the effect of covariates, a multiplicative model is used. These all methods have similar results.

• The Korean Journal of Applied Statistics
• /
• v.32 no.1
• /
• pp.69-81
• /
• 2019

We consider bivariate long range dependent (LRD) time series forecasting using a deep learning method. A long short-term memory (LSTM) network well-suited to time series data is applied to forecast bivariate time series; in addition, we compare the forecasting performance with bivariate fractional autoregressive integrated moving average (FARIMA) models. Out-of-sample forecasting errors are compared with various performance measures for functional MRI (fMRI) data and daily realized volatility data. The results show a subtle difference in the predicted values of the FIVARMA model and VARFIMA model. LSTM is computationally demanding due to hyper-parameter selection, but is more stable and the forecasting performance is competitively good to that of parametric long range dependent time series models.

• Journal of Environmental Impact Assessment
• /
• v.28 no.2
• /
• pp.83-100
• /
• 2019

This study, we conducted a non-parametric trend test (Seasonal Kendall tests, LOWESS) and Cross Correlation. We aimed to identify water quality trends using the weekly data for 9 variables (Water Temperature, EC, DO, BOD, COD, T-N, T-P, TOC and Chl-a) collected from 4 sites in the Paldang Lake from 2004.01 to 2016.12. According to the Seasonal Kendall test, Water temperature increased but EC, T-N and T-P decreased trend. LOWESS showed that BOD was gradually decreased from 2013 to 2016. but COD gradually increased between 2012 and 2016. As a result, it was confirmed that the period between 2012 and 2013 was a turning point in the increase of COD along with the decrease of BOD at all sites in Paldang Lake. Results of Cross Correlation showed that there was no time difference between all of Water variables and Sites. In this study, it is necessary to analyze the cause of the transition period and to monitoring the water quality more precisely for better water quality management in Paldang Lake.

• The Korean Journal of Financial Management
• /
• v.15 no.2
• /
• pp.279-337
• /
• 1998

이 논문에서는 주가가 확률과정, 즉 확률미분방정식에 의하여 생성되는가를 검정하고 주가의 운동법칙을 규명한다. 일별종합주가지수가 양수의 완전시계열상관을 갖고 있으며, 더욱이 3년 정도의 시차까지 의미있는 시계열상관을 갖고 있음이 발견되었다. 수익률과 가격변화의 시계열상관도 존재하고 시계열은 정상성(定常性)을 갖고 있다. 마팅게일에 의하여 주가가 생성되고있지 않음이 밝혀졌다. 한국증권거래소에서 계산하고 있는 일별 종합주가지수를 포함한 41개 산업별 지수를 사용하여 자본시장의 운동법칙을 규명하기 위하여 가장 많이 이용하고 있는 세개의 확률미분방정식을 검정하였다. 각 주가지수들이 온스타인 울렌벡 브라운 운동과정과 평균회귀과정을 따르지 않고 있다는 것이 발견되었다. 그러나 주가가 편류를 갖는 일반 기하 브라운 운동과정에 의하여 생성되고 있음이 검정을 통하여 확인되었다. 평균회귀과정에 의하여 주가가 생성되지 않는다는 발견은 의외라 할 수 있다. 주가가 온스타인 울렌벡 과정을 따르지 않는다는 것은 주가가 제 1계 정상적 자기회귀과정이 아니라는 것을 의미한다. 일별종합주가지수는 제 4계 자기회귀과정에 의하여 생성된다. 가격변화와 수익률의 생성함수는 제 4계 자기회귀과정이다. 종합주가지수의 제 1계 시계열상관계수는 1이다. 상당히 큰 시차를 갖을 때까지 시계열상관이 대략적으로 1을 유지하고 있다. 따라서 지수가 마팅게일을 따르고 있지 않다. 이 점은 가격변화와 수익률에 있어서도 유사하다. 가격변화, 수익률, 대수수익률의 제 1계 시계열상관이 0.1로 유의적이다. 따라서 수익도 마팅게일 과정을 따르고 있지 않다. 증권가격은 세 번에 걸쳐 구조의 번화가 발생하였다. 구조의 변화가 발생할 때마다 평균가격이 상승하였다. 이와 같은 현상은 장기적 기대가격이 미지일 가능성이 배제되지 않는다. 단기적 기대 주가가 알려진 반면 장기적 기대 주가가 미지라면 평균회귀과정은 장기적 기대주가로 회귀하고 있는 과정이므로 장기기대 주가의 미지성이 평균회귀 과정의 기각을 유도하게 된다. 우리나라의 투자자들은 무위험자산과 위험을 동시에 고려하여 투자활동을 전개하고 있음이 발견되었다. 선형의 효용함수를 갖는 위험중립적 태도의 투자자가 아니다. 위험기피형 효용함수 아래에서 투자활동을 수행하고 있는 합리적 투자자들이라 할 수 있다. 뿐 만 아니라 자신의 평생에 걸친 소비를 소비가 이루어지는 각 기마다 가급적 일정하게 하는 소비행동을 목표로 삼고 소비와 투자에 대한 의사결정을 내리고 있음이 실증분석을 통하여 밝혀졌다. 투자자들은 무위험 자산과 위험성 자산을 동시에 고려하여 포트폴리오를 구성하는 투자활동을 행동에 옮기고 있다.

• Journal of Soil and Groundwater Environment
• /
• v.11 no.2
• /
• pp.56-67
• /
• 2006

Trends of variation in groundwater levels, electrical conductivities and water temperatures obtained from the national groundwater monitoring stations (95 shallow and 169 deep wells) of Korea were evaluated. For the analysis, both parametric (linear regression) and non-parametric (Mann-Kendall test, Sen's test) methods were adopted. Results of linear regression analysis indicated that about 50% of the monitoring wells showed increasing trends of groundwater levels, electrical conductivities, and water temperatures and the others showed decreasing trends. However, the non-parametric analyses with monthly median values revealed that $14.8{\sim}20.0%$ of water levels were decreased, $24.2{\sim}36.9%$ of electrical conductivities were increased, and $27.4{\sim}32.5%$ of water temperatures were increased at a confidence level of 99%. Highly proportions of increasing or decreasing trends were unexpected and they resulted from the relatively short term of data collection (maximum 6 years). Meanwhile, the investigation of groundwater around the national groundwater monitoring stations showed that the decreasing or increasing trends of water levels, electrical conductivities, themselves, didn't indicate directly groundwater hazards such as groundwater depletion or groundwater contamination. Both the values and variation rates (slopes) of water level, electrical conductivity and temperature in the longer period are considered simultaneously. This study is the first comprehensive work in analyzing trends of groundwater data obtained from the national groundwater monitoring stations. Based on this study, the periodical and regular analysis of groundwater data is essentially required to grasp the overall variational trend of groundwater resources in the country.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2006.05a
• /
• pp.1460-1464
• /
• 2006

강수량의 특성 및 계절적인 양상은 지협적인 원인이기 보다는 해수면 온도(sea surface temperature)와 같은 기상 현상에 주로 영향을 받는다. 이러한 관점에서 강수량과 같은 수문변량의 장기적인 거동을 기상인자로부터 유추하고자 하는 연구는 무엇보다 중요하며 이러한 추론을 바탕으로 강수량의 장기예측 및 모의를 위한 기본적인 도구로 활용을 가능케 한다. 따라서 본 연구의 주요 목적은 해수면 온도를 기본으로 강수량과 기온의 변동성 및 상관성을 분석하고자 하며, 무엇보다 한반도 근해의 해수면 온도와의 직 간접적인 개연성을 살펴봄으로서 보다 효과적인 강수량 예측을 위한 하나의 변수로서의 가능성을 평가하고자 한다. 이를 위해 다양한 분석 방법 즉, 연주기를 제거하지 않은 자료의 선형적인 지체 상관 분석, 연주기를 제거하기 위해 표준화 된 자료의 지체 상관 분석 및 비모수적 상관분석을 수행하였다. 연주기를 제거하지 않은 자료의 경우 매우 강한 상관관계를 나타내었지만 이는 주로 계절 특성으로 인한 것으로 사료된다. 그러나 연주기를 제거한 Anomaly는 상대적으로 매우 작은 상관성을 보이고 있으나 유의성 검토를 통해 통계적으로 유의한 관계가 존재함을 확인 할 수 있었다. 따라서 강수량의 예측을 하나의 변수로서 이용이 가능할 것으로 사료되나 근해뿐만 아니라 한반도 기상의 연관성을 갖는 타 지역기상인자와의 보다 통합적인 검토가 필요하다 하겠다.

• Journal of the Korean Data and Information Science Society
• /
• v.21 no.6
• /
• pp.1237-1242
• /
• 2010

Based on the detailed travel history of cases from 2006 to 2008 who reside in non-malarious areas, statistical estimates of the incubation periods were obtained. The data suggest that cases fall into two categories with short- and long-term incubation periods, respectively. 72 and 25 cases successfully met our criteria for inferring the durations of short- and long-term incubation periods. The mean short- and long-term incubation periods were estimated to be 25.42 days and 328.6 days weeks, respectively.