• Journal of Information Processing Systems
• /
• 제20권3호
• /
• pp.345-359
• /
• 2024

A regression model represents the relationship between explanatory and response variables. In real life, explanatory variables often affect a response variable with a certain time lag, rather than immediately. For example, the marriage rate affects the birth rate with a time lag of 1 to 2 years. Although deep learning models have been successfully used to model various relationships, most of them do not consider the time lags between explanatory and response variables. Therefore, in this paper, we propose an extension of deep learning models, which automatically finds the time lags between explanatory and response variables. The proposed method finds out which of the past values of the explanatory variables minimize the error of the model, and uses the found values to determine the time lag between each explanatory variable and response variables. After determining the time lags between explanatory and response variables, the proposed method trains the deep learning model again by reflecting these time lags. Through various experiments applying the proposed method to a few deep learning models, we confirm that the proposed method can find a more accurate model whose error is reduced by more than 60% compared to the original model.

• 한국대기환경학회지
• /
• 제27권3호
• /
• pp.358-366
• /
• 2011

The purpose of this study was to analyze the monthly and seasonal PM10 data using the Autoregressive Error (ARE) model at the southern part of the Gyeonggi-Do, Pyeongtaek monitoring site in Korea. In the ARE model, six meteorological variables and four pollution variables are used as the explanatory variables. The six meteorological variables are daily maximum temperature, wind speed, amount of cloud, relative humidity, rainfall, and global radiation. The four air pollution variables are sulfur dioxide ($SO_2$), nitrogen dioxide ($NO_2$), carbon monoxide (CO), and ozone ($O_3$). The result shows that monthly ARE models explained about 17~49% of the PM10 concentration. However, the ARE model could be improved if we add the more explanatory variables in the model.

• 응용통계연구
• /
• 제34권4호
• /
• pp.589-597
• /
• 2021

단순회귀와 다중회귀에서 회귀계수의 의미는 차이가 있고 회귀계수의 추정값은 같지 않을 뿐 아니라 그 부호가 서로 다른 경우도 발생한다. 회귀모형에서 설명변수의 상대적 기여도의 파악은 회귀분석의 수행의 중요한 부분이다. 표준화 회귀모형에서 표준화 회귀계수는 해당 설명변수를 제외한 나머지 설명변수의 값이 고정되어있는 상황에서 설명변수가 표준편차만큼 증가하였을 때 반응변수가 표준편차를 기준으로 얼마나 변화했는가로 해석할 수 있지만 표준화 회귀계수의 크기가 각 설명변수의 상대적 중요도를 나타내는 척도라고 할 수 없음은 잘 알려져 있다. 본 논문에서는 다중회귀에서 회귀계수의 추정량을 상관계수와 결정계수의 함수로 나타내고 이를 추가적인 설명력과 추가적인 결정계수의 관점에서 생각해 본다. 또한 다양한 산점도에서의 상관계수와 회귀계수 추정값의 관계를 알아보고 설명변수가 두 개인 경우에 구체적으로 적용해 본다.

• 통상정보연구
• /
• 제11권2호
• /
• pp.111-126
• /
• 2009

This study analysed determinants of Foreign Direct Investment to ASEAN+ 3 member nations using panel data for which cross-sectional data are combined with time series data. The data for the analysis included the amount of FDI, GDP, and indexes of economic independence. This study collected data from six nations(Indonesia, Malaysia, Philippines, Singapore, Thailand, Vietnam) whose data were easily available, China and Japan from 2003 to 2007 and analysed them. The results are summarized as follows: Using the pooled OLS method, we found Model 2 had the highest explanatory power whose adjusted R-squared was 89.4%, which accounted for about 89% of foreign investment. Using the fixed effect model, Model 2 had the highest explanatory power whose adjusted R-squared was 96.8%, which accounted for about 97% of foreign investment. Using the probability effect model, Model 5 had the highest explanatory power, but in respect to its statistical significance, only GDP was 1% significant and the rest variables had no significance.

• Journal of the Korean Statistical Society
• /
• 제31권1호
• /
• pp.77-91
• /
• 2002

This paper considers a functional regression model with truncated errors in explanatory variables. We show that the ordinary least squares (OLS) estimators produce bias in regression parameter estimates under misspecified models with ignored errors in the explanatory variable measurements, and then propose methods for analyzing the functional model. Fully parametric frequentist approaches for analyzing the model are intractable and thus Bayesian methods are pursued using a Markov chain Monte Carlo (MCMC) sampling based approach. Necessary theories involved in modeling and computation are provided. Finally, a simulation study is given to illustrate and examine the proposed methods.

• Journal of the Korean Data and Information Science Society
• /
• 제20권6호
• /
• pp.1085-1092
• /
• 2009

지표오존 농도는 국가의 중요한 환경 척도 중의 하나이다. 본 연구에서는 경기도 파주시 오존농도를 자기회귀오차모형과 신경망모형으로 분석하였다. 오존 분석을 위한 설명변수로는 이산화황, 이산화질소, 일산화탄소, 프로메툼10 등의 대기자료와 일 최고온도, 풍속, 상대습도, 강수량, 이슬점온도, 운량, 수증기압 등의 기상자료를 사용하였다. 분석 결과 전반적으로 신경망모형이 좋은 모형으로 나타났고, 자기회귀오차모형도 오존에 영향을 주는 설명변수를 첨가하면 좋은 모형이 될 것으로 생각된다.

• 수산해양교육연구
• /
• 제27권4호
• /
• pp.998-1012
• /
• 2015

This study is smartphone addiction impulsiveness, stress, self-efficacy, and examine any changes to appear self-control. This study is a response to the results obtained for 310 people targeting high school in Pusan, the second grade students. For the analysis of the collected data by using the SPSS 22.0 program was the analysis of the T-test, ANOVA, Multiple Regression. The major findings of this study can be summed up as follows: first, smart phone addiction has significant difference in impulsivity, stress, self-efficacy, and self-control. Second, sex is found to be significant in impulsivity, stress, self-efficacy, and self-control. Third, grades are significant in impulsivity, self-efficacy, and self-control. Fourth, the model for impulsivity indicates 4% of explanatory power, which is significant. Fifth, explanatory power for stress is 4%, which is significant. Sixth, the model for self-efficacy shows 14% of explanatory power, which is significant. Meanwhile, smart phone addiction, sex, and grades have no significant effects on self-efficacy. Seventh, the model for self-control indicates 20% of explanatory power, which is significant.

• Journal of the Korean Data and Information Science Society
• /
• 제9권2호
• /
• pp.263-273
• /
• 1998

In this paper, we consider some approximate testings for the reliability of the stress-strength model when the stress X and strength Y each depends linearly on some explanatory variables z and w, respectively. We construct a bootstrap procedure for testing for various values of the reliability and compare the power of the bootstrap test with the test based on Mann-Whitney type estimator by Park et.al.(1996) for small and moderate sample size.

• 디지털융복합연구
• /
• 제19권5호
• /
• pp.241-247
• /
• 2021

본 연구는 1990년도부터 2019년까지 거래소에 상장된 기업들을 대상으로 발생액 추정모형의 설명력 변화를 분석하였다. 기존의 발생액 추정모형에 사용된 재무적 변수들이 특성이 시간이 지남에 따라 변화하거나, 전체 발생액 중에서 재량적 발생액의 비중이 변화하면 모형의 설명력에도 변화가 있을 것으로 기대하고 이를 가설화하여 분석하였다. 회귀 분석결과 수정 Jones 모형(1995)은 시간의 경과에 따라 그 설명력이 점차 낮아짐을 발견하였다. 이는 발생액 자체의 증가와 모형에 포함된 변수들의 분포가 변화함에 기인하는 것으로 추정된다. 본 연구의 시계열적 분석 결과는 이익조정 연구 등 학술적인 면이나 회계 정보를 이용하는 이용자에게 중요한 시사점을 제공할 것으로 기대된다.

• 한국도로학회논문집
• /
• 제14권5호
• /
• pp.189-199
• /
• 2012

PURPOSES: The purpose of this study is to propose a new methodology for developing statistical collision models and to show the validation results of the methodology. METHODS: A new modeling method of introducing variables into the model one by one in a multiplicative form is suggested. A method for choosing explanatory variables to be introduced into the model is explained. A method for determining functional forms for each explanatory variable is introduced as well as a parameter estimating procedure. A model selection method is also dealt with. Finally, the validation results is provided to demonstrate the efficacy of the final models developed using the method suggested in this study. RESULTS: According to the results of the validation for the total and injury collisions, the predictive powers of the models developed using the method suggested in this study were better than those of generalized linear models for the same data. CONCLUSIONS: Using the methodology suggested in this study, we could develop better statistical collision models having better predictive powers. This was because the methodology enabled us to find the relationships between dependant variable and each explanatory variable individually and to find the functional forms for the relationships which can be more likely non-linear.