In spatial data analysis, the prior conditional autoregressive (CAR) model is used to express the spatial dependence on random effects from adjacent regions. This paper provides a new proposed approach regarding the development of the existing normal CAR model into a more flexible, Fernandez–Steel skew normal (FSSN) CAR model. This approach is able to capture spatial random effects that have both symmetrical and asymmetrical patterns. The FSSN CAR model is built on the basis of the normal CAR with an additional skew parameter. The FSSN distribution is able to provide good estimates for symmetry with heavy- or light-tailed and skewed-right and skewed-left data. The effects of this approach are demonstrated by establishing the FSSN distribution and FSSN CAR model in spatial data using Stan language. On the basis of the plot of the estimation results and histogram of the model error, the FSSN CAR model was shown to behave better than both models without a spatial effect and with the normal CAR model. Moreover, the smallest widely applicable information criterion (WAIC) and leave-one-out (LOO) statistical values also validate the model, as FSSN CAR is shown to be the best model used.
Data with a multimodal pattern can be analyzed using a mixture model. In a mixture model, the most important step is the determination of the number of mixture components, because finding the correct number of mixture components will reduce the error of the resulting model. In a Bayesian analysis, one method that can be used to determine the number of mixture components is the reversible jump Markov chain Monte Carlo (RJMCMC). The RJMCMC is used for distributions that have location and scale parameters or location-scale distribution, such as the Gaussian distribution family. In this research, we added an important step before beginning to use the RJMCMC method, namely the modification of the analyzed distribution into location-scale distribution. We called this the non-Gaussian RJMCMC (NG-RJMCMC) algorithm. The following steps are the same as for the RJMCMC. In this study, we applied it to the Weibull distribution. This will help many researchers in the field of survival analysis since most of the survival time distribution is Weibull. We transformed the Weibull distribution into a location-scale distribution, which is the extreme value (EV) type 1 (Gumbel-type for minima) distribution. Thus, for the mixture analysis, we call this EV-I mixture distribution. Based on the simulation results, we can conclude that the accuracy level is at minimum 95%. We also applied the EV-I mixture distribution and compared it with the Gaussian mixture distribution for enzyme, acidity, and galaxy datasets. Based on the Kullback–Leibler divergence (KLD) and visual observation, the EV-I mixture distribution has higher coverage than the Gaussian mixture distribution. We also applied it to our dengue hemorrhagic fever (DHF) data from eastern Surabaya, East Java, Indonesia. The estimation results show that the number of mixture components in the data is four; we also obtained the estimation results of the other parameters and labels for each observation. Based on the Kullback–Leibler divergence (KLD) and visual observation, for our data, the EV-I mixture distribution offers better coverage than the Gaussian mixture distribution.
1 February 2021 in Journal of Physics: Conference Series
There is a lot of research on infectious diseases. One of the dangers is caused by dengue hemorrhagic fever (DHF) viruses. Almost all the world there are dengue viruses that spread. In health science, this research continues to increase over time. It has been widely studied by medical circles, but from a statistical point of view, patient survival can be investigated probabilistically. The survival time of DHF patients could refer to the time of arrival of DHF patients to the hospital with all the health conditions of the patient to be discharged from the hospital. It is because the patient has been declared medically cured or further improved from the initial hospital admission. In this case, survival analysis will be appropriate to be implemented, where the event of the study is DHF patients who are discharged from the hospital due to recovery. DHF patient survival data is found following the multimodal Generalized Extreme Value distribution so that a mixture model would be employed. There is a lot of transmission media for diseases transmitted by the DHF virus, but the fastest transmission medium occurs in a short time, called the Aedes Aegypti Mosquito. This transmission of the DHF virus by mosquitoes can occur between adjacent neighboring regions. This article brings together two fields of research, namely survival models with spatial random effects. The Bayesian analysis will be used as a model parameter estimation approach, whereby unmeasured autocorrelations between intersecting regions will be captured through the conditional autoregressive (CAR). Two different distributions approach for these unmeasured autocorrelations will be given, i.e. the Normal CAR and the Double Exponential CAR. Four predictor variables, namely sex, age, hematocrit level, and platelet count in each DHF patient are used to explain the variability of the survival. The Cox Proportional Hazard regression is employed to find out the significant influence of those variables on the survival spatial DHF model. The result shows that the two-component mixture Generalized Extreme Value regression model coupled with a random Double Exponential CAR effect is the best model. Based on the best model obtained, the significant predictor variables for the DHF spatial survival model are sex, hematocrit level, and platelet count.
Dengue Hemorrhagic Fever (DHF) is one of the most common diseases experienced in Indonesia, especially in the rainy season. DHF is caused by the dengue virus which is transmitted by female Aedes aegypti mosquitoes. The recovery rate of DHF patients is influenced by several factors. In this study, we are modeling how these factors can influence the recovery rate for DHF patients. Thus, we can provide advice to the hospital. Survival analysis is a method used to model survival data. In this case, the survival model will be used to find the relationship between the recovery rate of DHF patients and factors that can affect it. The environment of patient residence is one of the factors that need to be considered. This case required analysis by considering the spatial effect of the proximity between one sub-district and another. The addition of a random effect can be done to avoid bias on estimation by using conditionally autoregressive (CAR) prior. The result showed that DHF in eastern Surabaya mostly occurred in Tambaksari and Gubeng sub-districts. The mean of survival time of patients is 4 days. In this study, the Weibull and Generalized Gamma distributions were used to modeling the patient recovery rate. In this study, there are spatial autocorrelations cases which are measured with the Moran's I index. The Weibull distribution with addition frailty using Normal CAR is the best model that can be used to modeling the recovery rate of DHF patients.
1 January 2021 in Journal of Physics: Conference Series
In statistics, there are many types of data. Some data carry information about the location where observations occur, so that they can have a spatial effect. Dengue hemorrhagic fever (DHF) data which is easily transmitted, will be consequently has a spatial effect on its patient survival. In this study, we included DHF patient recovery time as a response variable, and several other variables as covariates considered to influence the patient’s recovery time. Our aim in this study is to model how these variables affect the recovery rate for DHF patients with the accompanying patient residence as the spatial effects. Survival analysis is the best method for modeling the recovery rate for DHF patients. A conditional autoregressive (CAR) model is given to explain the relationship between adjacent locations, which is not explained in the general survival analysis. Several researchers have used the Cox model coupled with the Normal CAR. In this study, we used the Cox model using Normal CAR and compared it with the Double-Exponential (DE) CAR. To estimate the regression parameters of the Cox model, we used the Stan software. The advantage of Stan compared to the other Bayesian software such as BUGS and JAGS is the creativity of the researcher in writing the distribution as user-defined, as well as writing the CAR model in the Stan. Based on the WAIC value, modeling the DHF data using the Cox model coupled with the DE CAR is better than coupled with the Normal CAR. Based on the best model, variables that affect the recovery rate of DHF patients are age, the high schools in last education, unemployed in the type of occupations, the stadium II in severity level, pulse, temperature, and leukocytes.
Introduction: Motor vehicle theft poses significant challenges in urban areas, particularly in large metropolitan cities like Surabaya, Indonesia's second-largest city. Surabaya's strategic economic role makes it a hotspot for criminal activities, including motor vehicle theft, driven by various socio-economic factors.Methods: This study utilizes the Space-Time Autoregressive Integrated Moving Average (STARIMA) model to predict motor vehicle theft cases across five sub-regions in Surabaya, covering the period from January 2019 to December 2023. The STARIMA model, which incorporates both temporal and spatial dependencies, offers a more robust framework for crime prediction compared to traditional models like ARIMA. The results show that STARIMA effectively captures the spatio-temporal dynamics of crime, providing valuable insights for law enforcement to develop targeted strategies that enhance public safety.Results: The model's performance was evaluated using the Root Mean Square Error (RMSE), indicating its suitability for accurate and actionable crime forecasting in Surabaya. Based on the RMSE value, the best model obtained is STARIMA (1,1,2) with a Uniform Location weighting matrix.Conclusions: This STARIMA (1,1,2) model, it is used to predict motor vehicle theft incidents in West Surabaya, Central Surabaya, South Surabaya, East Surabaya, and North Surabaya. The forecast value carried out is for a period of five months into the future. Case predictions for the next five months show fluctuations in each region of Surabaya, with the highest regions in succession being North Surabaya, East Surabaya, and South Surabaya