• 응용통계연구
• /
• 제35권5호
• /
• pp.631-644
• /
• 2022

본 논문은 고차원 시계열 자료에 이상점이 존재하는 경우 희박벡터자기회귀모형(sparse VAR; sVAR)의 모수를 강건하게 추정하는 방법에 대해서 연구하였다. 먼저 Xu 등 (2008)이 독립인 자료에서 밝혔듯이 adaptive lasso 방법이 sVAR 모형에서도 어느 정도의 강건함을 가짐을 모의 실험을 통해 알 수 있었다. 하지만, 이상점의 개수가 증가하거나 이상점의 영향력이 커지는 경우 효율성이 현저히 저하되는 현상도 관찰할 수 있었다. 따라서 이를 개선하기 위해서 최소절대편차(least absolute deviation; LAD)와 Huber 함수를 기반으로 벌점화 시키는 adaptive lasso를 이용하여 sVAR 모형을 추정하는 방법을 본 논문에서는 제안하고 그 성능을 검토하였다. 모의 실험을 통해 제안한 로버스트 추정 방법이 이상점이 존재하는 경우에 모수 추정을 더 정확하게 하고 예측 성능도 뛰어남을 확인했다. 또한 해당 방법론들을 전력사용량 데이터에 적용한 결과 이상점으로 의심되는 시점들이 존재하였고, 이를 고려하여 강건하게 추정하는 제안한 방법론이 더 좋은 예측 성능을 보임을 확인할 수 있었다.

• Journal of the Korean Data and Information Science Society
• /
• 제22권5호
• /
• pp.839-847
• /
• 2011

이 연구에서는 다중 선형회귀 모형에서 자기공분산에 근거한 회귀 계수의 추정량을 도출하였다. 자기공분산에 근거한 방법은 Park (2009)에 제시된 방법으로 직관적으로 매혹적이지는 않지만, 이것에 근거한 추정량이 회귀 계수의 불편추정량이 된다. 설명변수 벡터가 어떤 정칙조건을 만족한다면, 오차가 자기회귀이동평균 모형을 따르면 만족되는 약한 조건 하에서 이 추정량이 최소제곱 추정량과 점근적으로 동일한 분포를 가지며 또한 회귀 계수에 확률 상 수렴한다는 것을 보였다. 마지막으로 모의실험을 통해 이 성질들이 소표본에서도 성립하는 것을 보였다.

• 지식경영연구
• /
• 제16권4호
• /
• pp.109-132
• /
• 2015

We examine offline and online channel sales of experience goods, and compare and contrast the sales patterns of existing products and new products between channels. To this end, we obtain the channel-specific time-series sales data from the leading company selling beauty products, both offline and online. By applying the Vector Autoregressive Model, we empirically find out how the relationship between existing products and new products changes between the shopping channels. Our empirical findings are as follows. First, the sales effects from existing products to new products are significantly positive at both offline and online channels, and this positive effect is greater in the offline channel than in the online channel. Second, the influence of new products on existing products is more positive in the offline channel than in the online channel. Third, the impact of existing products sales on new products sales is greater than that of new products on existing products. Lastly, the inertia effect, the effect within the same shopping channel and the same selling product, is significantly positive in the offline channel but not in the online channel, and this asymmetric inertia effect emerges as we focus on experience goods. Moreover, the impulse response function analysis provides the three important implications. First, companies should pay attention to the same channel but different types of products. Second, the offline channel is more vulnerable to market shock than the online channel. Third, new products sales vary by existing products sales to the greater extent, compared to the opposite relationship. We believe our study contributes theoretically and practically to the fields of marketing and knowledge management.

• 경제분석
• /
• 제25권3호
• /
• pp.100-137
• /
• 2019

본 연구에서는 요인활용 다변수 자기회귀모형(FAVAR)과 2000년 이후 한국의 167개 거시변수를 이용하여 정부지출 증가가 거시변수에 미치는 영향 및 각 산업에 미치는 영향을 분석하였다. 정부지출의 영향을 정부소비지출과 정부투자지출로 나누어 추정한 결과 두 형태의 지출의 효과가 크게 다른 것으로 나타나 이를 고려하지 않는 경우 정확한 정부지출의 효과를 분석하는 것이 어려울 수 있음을 보였다. 특히 정부소비지출은 1년여의 시차를 두고 경기부양효과가 비교적 뚜렷하나, 정부투자지출의 경우 뚜렷한 경기부양효과를 찾아보기 어려웠다. 또한 전통적인 민간소비나 민간투자를 구축하는 채널보다는 수입수요의 증가로 인한 순수출 감소가 재정지출, 특히 정부소비지출의 승수효과를 감소시키는 것으로 보인다. 산업별로는 두 가지 형태의 정부지출증가 모두 토목건설업에 긍정적인 영향을 미치는 것으로 나타났으며, 정부소비지출의 증가는 대부분의 제조업과 서비스업에도 부양효과가 있는 것으로 나타났다.

• 한국산림과학회지
• /
• 제96권2호
• /
• pp.222-226
• /
• 2007

This study developed the Korean roundwood import prediction model using vector autoregressive (VAR) method. The roundwood was divided into softwood and hardwood by species. The VAR model of roundwood import was specified with two lagged endogenous variables, that is, roundwood import volume and roundwood import price. The results showed that the significance levels of F-statistics in the softwood and hardwood roundwood import equations rejected the hypothesis that all coefficients are zero. So, we concluded that roundwood import volume can be explained by lagged import volume and lagged import price in Korea. The coefficient signs of all variables were as expected. Also, the model has good explanatory power, and there is no autocorrelation.

• The Journal of Asian Finance, Economics and Business
• /
• 제9권1호
• /
• pp.189-201
• /
• 2022

The study empirically examines the impact of monetary fundamentals along with global oil prices on the Pak-rupee exchange rate using the monthly data over 2001-2020. Employing the cointegrating vector autoregressive with exogenous variables (VARX) and vector error correction model with exogenous variables (VECMX), the study analyzes the impact of domestic monetary fundamentals while considering the foreign variables as weakly exogenous. In order to account for the structural breaks in the data, the Lagrange multiplier (LM) unit root test with two structural breaks has been used (Lee & Strazicich, 2003). The empirical results reveal that the domestic and foreign monetary variables significantly explain the exchange rate movements in Pakistan both in the long run and in the short run. The dynamic properties of the monetary model of exchange rate have been analyzed using the persistence profile analysis and generalized impulse response functions (GIRFs). The results reveal that the responses of shocks to domestic monetary fundamentals are consistent with the predictions of the monetary model of the exchange rate. Furthermore, being a net oil importer, a rise in global oil prices significantly depreciated the Pak-rupee exchange rate over the period of study. The global financial crisis (GFC) and pandemic (COVID-19) were also found to cause the Pak-rupee exchange rate depreciation.

• 한국소음진동공학회논문집
• /
• 제13권1호
• /
• pp.48-55
• /
• 2003

Hidden Markov Model(HMM) has a doubly embedded stochastic process with an underlying stochastic process that can be observed through another set of stochastic processes. This structure of HMM is useful for modeling vector sequence that doesn't look like a stochastic process but has a hidden stochastic process. So, HMM approach has become popular in various areas in last decade. The increasing popularity of HMM is based on two facts : rich mathematical structure and proven accuracy on critical application. In this paper, we applied continuous HMM (CHMM) approach with AR coefficient to detect and predict the chatter of lathe bite and to diagnose the wear of oil Journal bearing using rotor shaft displacement. Our examples show that CHMM approach is very efficient method for machine health monitoring and prediction.

• Communications for Statistical Applications and Methods
• /
• 제22권6호
• /
• pp.615-624
• /
• 2015

We consider the estimation of seasonal cointegration in the presence of conditional heteroskedasticity (CH) using a feasible generalized least squares method. We capture cointegrating relationships and time-varying volatility for long-run and short-run dynamics in the same model. This procedure can be easily implemented using common methods such as ordinary least squares and generalized least squares. The maximum likelihood (ML) estimation method is computationally difficult and may not be feasible for larger models. The simulation results indicate that the proposed method is superior to the ML method when CH exists. In order to illustrate the proposed method, an empirical example is presented to model a seasonally cointegrated times series under CH.

• 통상정보연구
• /
• 제12권4호
• /
• pp.171-192
• /
• 2010

In general, a stock index and its individual stocks are assumed to follow a random walk. A stock index is an important source of information and one that is seen by people everyday, regardless of their investment intentions. This paper examines the correlation between the KOSPI-the index that best reflects the Korean stock market and the macro - economic variables that have been found to influence the index by previous studies. The sample period considers the years after 2000 when the Korean stock market matured as restrictions on foreign investors were removed. For this purpose, a Vector Error Correction Model (VECM) and KOSPI equation with a general pacific approach were used. This paper aims at verifying the factors that determined the KOSPI after 2000 and at examining whether there was structural change in the investment environment. It also investigates changes in the factors determining the KOSPI's performance as a result of structural changes in the investment environment. The V AR (Vector Autoregressive) model including the nine variables was selected as a baseline model whose stability was tested using the unit root test. The results from the VECM and the structural changes in the investment environment can be summarized by the following Inner story points.

• East Asian Economic Review
• /
• 제20권4호
• /
• pp.519-544
• /
• 2016

This paper examines quantile dependence and directional predictability between the foreign exchange market and the stock market in Korea. Instead of adopting a multivariate model such as a vector autoregressive model, a multivariate GARCH model or a combination of both models, we apply the cross-quantilogram recently proposed by Han et al. (2016). Considering various quantile ranges, we investigate various spillover effects between two markets. Our findings show that there exists an asymmetric bi-directional spillover between two markets and the interdependence between two markets implies that one market has significant predictive power on the other.