• 한국건설관리학회논문집
• /
• 제25권4호
• /
• pp.53-65
• /
• 2024

During COVID-19 the housing market in Korea experienced the soaring prices, despite the decrease in the economic growth rate. This paper aims to analyze macroeconomic determinants affecting housing prices in Korea during the pandemic and find an appropriate statistic model to forecast the changes in housing prices in Korea. First, an appropriate lag for the model using Akaike information criterion was found. After the macroeconomic factors were checked if they possess the unit root, the dependencies in the model were analyzed using vector autoregression (VAR) model. As for the prediction, the VAR model was used and, besides, compared afterwards with the long short-term memory (LSTM) model. CPI, mortgage rate, IIP at lag 1 and federal funds effective rate at lag 1 and 2 were found to be significant for housing prices. In addition, the prediction performance of the LSTM model appeared to be more accurate in comparison with the VAR model. The results of the analysis play an essential role in policymaker perception when making decisions related to managing potential housing risks arose during crises. It is essential to take into considerations macroeconomic factors besides the taxes and housing policy amendments and use an appropriate model for prices forecast.

• Ocean and Polar Research
• /
• 제34권2호
• /
• pp.185-195
• /
• 2012

This paper focused on forecasting a short-term production of oysters, which have been farmed in Korea, with distinct periodicity of production by year, and different production level by month. To forecast a short-term oyster production, this paper uses monthly data (260 observations) from January 1990 to August 2011, and also adopts several econometrics methods, such as Multiple Regression Analysis Model (MRAM), Seasonal Autoregressive Integrated Moving Average (SARIMA) Model, and Vector Error Correction Model (VECM). As a result, first, the amount of short-term oyster production forecasted by the multiple regression analysis model was 1,337 ton with prediction error of 246 ton. Secondly, the amount of oyster production of the SARIMA I and II models was forecasted as 12,423 ton and 12,442 ton with prediction error of 11,404 ton and 11,423 ton, respectively. Thirdly, the amount of oyster production based on the VECM was estimated as 10,425 ton with prediction errors of 9,406 ton. In conclusion, based on Theil inequality coefficient criterion, short-term prediction of oyster by the VECM exhibited a better fit than ones by the SARIMA I and II models and Multiple Regression Analysis Model.

• 대기
• /
• 제24권4호
• /
• pp.503-514
• /
• 2014

Evaluation of predictability of numerical models for tropical cyclone track was performed using along-and cross-track component. The along-and cross-track bias were useful indicators that show the numerical models predictability associated with cause of errors. Since forecast errors, standard deviation and consistency index of along-track component were greater than those of cross-track component, there was some rooms for improvement in alongtrack component. There was an overall slow bias. The most accurate model was JGSM for 24-hour forecast and ECMWF for 48~96-hour forecast in direct position error, along-track error and cross-track error. ECMWF and GFS had a high variability for 24-hour forecast. The results of predictability by track type showed that most significant errors of tropical cyclone track forecast were caused by the failure to estimate the recurvature phenomenon.

• 대기
• /
• 제29권2호
• /
• pp.219-226
• /
• 2019

Generally, the weather forecast system has been run using prescribed ocean condition. As it is widely known that coupling between atmosphere and ocean process produces consistent initial condition at all-time scales to improve forecast skill, there are many trials on the application of data assimilation of coupled model. In this study, we implemented a weakly coupled data assimilation (short for WCDA) system in global NWP model with low horizontal resolution for coupled forecast with uncoupled initialization, following WCDA system at the Met Office. The experiment is carried out for a typhoon evolution forecast in 2017. Air-sea exchange process provides SST cooling and gives a substantial impact on tendency of central pressure changes in the decaying phase of the typhoon, except the underestimated central pressure. Coupled data assimilation is a challenging new area, requiring further work, but it would offer the potential for improving air-sea feedback process on NWP timescales and finally contributing forecast accuracy.

• 한국지리정보학회지
• /
• 제2권2호
• /
• pp.69-78
• /
• 1999

본 논문에서는 한반도의 지역별 강수량 예측을 위한 신경망 기법을 소개한다. 시계열 패턴 예측 문제에 적용될 수 있는 기존의 다양한 신경망 모델의 특성을 분석하고 이로부터 강수량 예측 문제에 적합한 모델 및 학습 알고리즘을 제시한다. 본 논문에서 제시하는 모델은 계층적구조의 신경망으로 각 노드의 출력값은 일정기간동안 버퍼에 저장되어 상위계층에 입력으로 작용한다. 본 연구에서는 제안된 모델에 대하여 이중연결형태의 시냅스 구조를 채택하고, 이에 대한 네트워크의 동작특성과 학습알고리즘 등을 정의한다. 이러한 이중연결구조는 기존의 다층퍼셉트론에서 바이어스 노드의 역할을 담당하며, 노드가 갖는 특징들간의 관계를 효과적으로 반영함으로써 기존의 전형적인 시계열 예측 신경망인 FIR(Finite Impulse Response) 네트워크와 비교할 때 학습의 효율을 개선시킨다. 제시된 이론은 월별 및 계절별 강수량 예측 실험에 적용하였다. 신경망 예측기의 학습자료로서 과거 수십년동안 관측된 강수량 데이터와 해수표면온도 데이터를 사용하며 예측 실험결과로부터 제시된 이론의 타당성을 고찰한다.

• Structural Engineering and Mechanics
• /
• 제75권4호
• /
• pp.477-485
• /
• 2020

Finite element (FE) model based structural damage detection (SDD) methods play vital roles in effectively locating and quantifying structural damages. Among these methods, structural model updating should be conducted before SDD to obtain benchmark models of real structures. However, the characteristics of updating parameters are not reasonably considered in existing studies. Inspired by the l_∞ norm regularization, a novel anti-sparse representation method is proposed for structural model updating in this study. Based on sensitivity analysis, both frequencies and mode shapes are used to define an objective function at first. Then, by adding l_∞ norm penalty, an optimization problem is established for structural model updating. As a result, the optimization problem can be solved by the fast iterative shrinkage thresholding algorithm (FISTA). Moreover, comparative studies with classical regularization strategy, i.e. the l₂ norm regularization method, are conducted as well. To intuitively illustrate the effectiveness of the proposed method, a 2-DOF spring-mass model is taken as an example in numerical simulations. The updating results show that the proposed method has a good robustness to measurement noises. Finally, to further verify the applicability of the proposed method, a six-storey aluminum alloy frame is designed and fabricated in laboratory. The added mass on each storey is taken as updating parameter. The updating results provide a good agreement with the true values, which indicates that the proposed method can effectively update the model parameters with a high accuracy.

• 한국환경과학회지
• /
• 제24권7호
• /
• pp.865-874
• /
• 2015

This study was conducted to investigate the correlation between the distribution chart and input data of the predicted 3-second gust and damage cost, by using the forecast field and analysis field of Regional Data Assimilation Prediction System (RDAPS) as initial input data of Korea risk assessment model (RAM) developed in the preceding study. In this study the cases of typhoon Rusa which caused occurred great damage to the Korean peninsula was analyzed to assess the suitability of initial input data. As a result, this study has found out that the distribution chart from the forecast field and analysis field predicted from the point where the effect due to the typhoon began had similarity in both 3-second gust and damage cost with the course of time. As a result of examining the correlation, the 3-second gust had over 0.8, and it means that the forecast field and analysis field show similar results. This study has shown that utilizing the forecast field as initial input data of Korea RAM could suit the purpose of pre-disaster prevention.

• 한국건설관리학회논문집
• /
• 제8권4호
• /
• pp.203-213
• /
• 2007

In the past, several construction projects have exceeded their schedule resulting in financial losses to the owners; at present there are very few methods available to accurately forecast the completion date of a project. These nay be because of unforeseen outcomes that cannot be accounted for earlier and because of deficiency of proper tools to forecast completion date of said project. To overcome these difficulties, project managers may need a tool to predict the completion date at the early stage of project development. Bayesian Inference introduced in this paper is one such tool that can be employed to forecast project progress at all construction stages. Using this inference, project managers can combine an initially planned project progress (growth curve) with reported information from ongoing projects during the development, and in addition, dynamically revise this initial plan and quantify the uncertainty of completion date. This study introduces a theoretical model and proposes a mathematically information-based framework to forecast a project completion date that corresponds with the actual progress data and to monitor the modified uncertainties using Bayesian Inference.

• 한국건축시공학회:학술대회논문집
• /
• 한국건축시공학회 2014년도 춘계 학술논문 발표대회
• /
• pp.14-15
• /
• 2014

The modular construction is as yet early stage of market in Korea. So It is have difficulty of market demand forecast of the modular building. Therefore, this study was done analysis for market trends of the modular building using ARIMA(Auto Regressive Integrated Moving Average) model by time series data.

• Communications for Statistical Applications and Methods
• /
• 제12권2호
• /
• pp.295-304
• /
• 2005

This paper concentrates on the prediction of typhoon tracks using the dynamic linear model (DLM) for the statistical correction of the numerical model guidance used in the JMA. The DLM with proposed forecast strategy is applied to reduce their systematic errors using the latest observation. All parameters of the DLM are updated dynamically and backward forecasting is performed to remove the effect of initial values.