• Proceedings of the Korean Statistical Society Conference
• /
• 2003.05a
• /
• pp.269-273
• /
• 2003

We consider an efficient parametric estimation method of spatial dependence in weak stationary processes. Spatial dependence is modeled through variogram and correlogram. Most of parametric estimation methods of correlogram use two step method; nonparametric estimation and parametric integration. We bind these two steps into one step by using GEE method instead of least squares type optimization. Our one step method is more efficient statistically and gives a clear interpretation of related concepts used in traditional two step methods.

• The Korean Journal of Applied Statistics
• /
• v.17 no.3
• /
• pp.519-533
• /
• 2004

In this paper, we suggest quasi-likelihood estimation (QLE) method and its robust version in estimating spatial dependence modelled through variogram used for spatial data modelling. We compare the statistical characteristics of the estimators with other popular least squares estimators of parameters for variogram model by simulation study. The QLE method for estimating spatial dependence has the advantages that it does not need the concept of lags commonly required for least squares estimation methods as well as its statistical superiority. The QLE method also shows the statistical superiority to the other methods for the tested Gaussian and non-Gaussian spatial processes.

• Journal of Korea Planning Association
• /
• v.54 no.7
• /
• pp.89-102
• /
• 2019

The growth of vacant housing has been problematic in both Korea and other countries as it causes various socio-economic problems and negatively affects residential environments. Despite the importance of effectively managing vacant housing, few studies have been undertaken regarding spatial patterns of vacant housing growth. This study aims to examine spatial dependence in vacant housing growth. We used 2005 and 2015 Population and Housing Census and employed spatial modeling. The empirical analysis shows that there is spatial dependence in vacant housing growth. Also, the spatial clusters of growing vacant housing are present in the non-capital region and nearby cities while the spatial clusters of declining vacant housing are present in the capital region. The policy implications of this study are as follows: First, local governments should make collaborate efforts with geographically proximate cities for more effective management of vacant housing. Second, given that vacant housing is more prevalent and growing in the non-capital region, it is necessary to employ differential policies to manage housing vacancy between the capital and non-capital regions.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2015.05a
• /
• pp.180-180
• /
• 2015

Weather radar has been widely used in measuring precipitation and discharge and predicting flood risks. The radar rainfall estimate has one of the essential problems in terms of uncertainty and accuracy. Previous study analyzed radar errors to reduce its uncertainty or to improve its accuracy. Furthermore, a recent analyzed the effect of radar error on rainfall-runoff using spatial error model (SEM). SEM appropriately reproduced radar error including spatial correlation. Since the SEM does not take the time dependence into account, its time variability was not properly investigated. Therefore, in the current study, we extend the SEM including time dependence as well as spatial dependence, named after Spatial-Temporal Error Model (STEM). Radar rainfall events generated with STEM were tested so that the peak runoff from the response of a basin could be investigated according to dependent error. The Nam River basin, South Korea, was employed to illustrate the effects of STEM on runoff peak flow.

• Journal of Biosystems Engineering
• /
• v.29 no.2
• /
• pp.109-120
• /
• 2004

To obtain basic information for precision agriculture, spatial variability of rice growth condition was evaluated in 100m ${\times}$100m paddy field. The rice growth condition of four hundred locations in the field were investigated to analyze the spatial variability of their properties ; SPAD, plant length and tiller number. Geostatistical analysis was carried out to examine within-field spatial variability using semivariograms and kriged maps as well as descriptive statistics. Descriptive statistics showed that the coefficient of variation for SPAD, plant length, and tiller number exceeded 5.70 %, suggesting a relatively high variability. Geostatistical analysis indicated a high spatial dependence for all the properties except for the second tiller number. The range of spatial dependence was about 20 m for SPAD, plant length, and tiller number. Based on the results of spatial dependence, kriged maps were prepared for the properties to analyse their spatial distribution in the field. The results reflected the history of field management. In conclusion, the need for site-specific field management and possibility of precision agriculture were demonstrated even in an almost flat paddy field.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.5
• /
• pp.1225-1239
• /
• 2016

We investigate pediatric tumor incidence data collected by the Florida Association for Pediatric Tumor program using various models commonly used in disease mapping analysis. Particularly, we consider Poisson normal models with various conditional autoregressive structure for spatial dependence, a zero-in ated component to capture excess zero counts and a spatio-temporal model to capture spatial and temporal dependence, together. We found that intrinsic conditional autoregressive model provides the smallest Deviance Information Criterion (DIC) among the models when only spatial dependence is considered. On the other hand, adding an autoregressive structure over time decreases DIC over the model without time dependence component. We adopt weighted ranks squared error loss to identify high risk regions which provides similar results with other researchers who have worked on the same data set (e.g. Zhang et al., 2014; Wang and Rodriguez, 2014). Our results, thus, provide additional statistical support on those identied high risk regions discovered by the other researchers.

• Asia-Pacific Journal of Business
• /
• v.13 no.1
• /
• pp.277-288
• /
• 2022

Purpose - This study analyzed the determinants of local unemployment rate in Korea using panel data from 16 metropolitan cities and provinces from 2000 to 2018. Design/methodology/approach - We use a dynamic spatial panel model that considers characteristics of the regional unemployment rate such as the common factors effect, spatial dependence, and serial correlations. Findings - The local unemployment rate is affected by the past and present values of the national unemployment rate. And it is significantly affected by the past local unemployment rate and the past neighboring unemployment rate because spatial dependence and serial correlations are clearly present. In addition, when the industrial structure diversity and labor productivity were high, the regional unemployment rate decreased, and when the education level was high, the regional unemployment rate increased. Research implications or Originality - In order to reduce regional unemployment rate, it is necessary to plan and establish regional customized industrial structure policies under the stance of diversification rather than specializing the regional industrial structure and accompany improvement of the quality of education with the number of years of education. In addition, the redistribution of labor from low labor productivity sectors to high labor productivity sectors through technology development will help to reduce the local unemployment rate.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.5
• /
• pp.1909-1928
• /
• 2020

Predicting human mobility has always been an important task in Location-based Social Network. Previous efforts fail to capture spatial dependence effectively, mainly reflected in weakening the location topology information. In this paper, we propose a neural network-based method which can capture spatial-temporal dependence to predict the next location of a person. Specifically, we involve a graph convolutional network (GCN) based on a seq2seq framework to capture the location topology information and temporal dependence, respectively. The encoder of the seq2seq framework first generates the hidden state and cell state of the historical trajectories. The GCN is then used to generate graph embeddings of the location topology graph. Finally, we predict future trajectories by aggregated temporal dependence and graph embeddings in the decoder. For evaluation, we leverage two real-world datasets, Foursquare and Gowalla. The experimental results demonstrate that our model has a better performance than the compared models.

• The Korean Journal of Applied Statistics
• /
• v.30 no.2
• /
• pp.215-231
• /
• 2017

The US presidential primaries take place sequentially in different places with a time lag. However, they have not attracted as much attention in terms of modelling as the US presidential election has. This study applied several autologistic models to find the relation between the outcome of the primary election for a Democrat candidate with socioeconomic attributes in consideration of spatial and temporal dependence. According to the result applied to the 2016 election data at the county level, Hillary Clinton was supported by people in counties with high population rates of old age, Black, female and Hispanic. In addition, spatial dependence was observed, representing that people were likely to support the same candidate who was supported from neighboring counties. Positive auto-correlation was also observed in the time-series of the election outcome. Among several autologistic models of this study, the model specifying the effect of Super Tuesday had the best fit.

• Journal of the Architectural Institute of Korea Planning & Design
• /
• v.36 no.5
• /
• pp.23-32
• /
• 2020

Recently, women's fear of crime continues to increase in space of everyday. By the way, the fear of crime has the spatial properties as crime. Therefore, The purpose of this study is to evaluate the spatial dependence of fear of crime and to suggest the physical environmental factors influencing fear of crime. For this, a spatial regression analysis using spatial weights was conducted based on the location data of the fear of crime measured through a survey. The results of this study are as follows; First, the fear of crime felt by women in their twenties who are vulnerable to crime has spatial dependence. Therefore, it is necessary to consider the spatial characteristics in analyzing the environmental factors affecting this. Second, in order to reduce the fear of crime, it is necessary to improve the environments of old housing and entertainment facilities. There is also a need for ongoing management. Third, careful consideration is needed in the installation of CCTV and street lights, which are factors influencing the fear of crime. It is necessary to establish a reasonable arrangement standard for CCTV and to analyze the street lighting in detail.