RISK ASSESSMENT OF STOCKS PORTFOLIO THROUGH ENSEMBLE ARMA-GARCH AND VALUE AT RISK (CASE STUDY: INDF.JK AND ICBP.JK STOCK PRICE)

Stocks portfolio is a form of investment that can be used to minimize the risk of loss. In a stock portfolio, the Value at Risk (VaR) can be predicted through the portfolio return. If portfolio return variance is heteroskedastic risk prediction can be done by using VaR with ARIMA-GARCH or Ensemble ARIMA-GARCH model approach. Furthermore, the accuracy of VaR is tested through Backtesting test. In this study, the portfolio is formed from PT Indofood CBP Sukses Makmur (ICBP.JK) and PT Indofood Sukses Makmur Tbk (INDF.JK) stocks from 01/01/2018 to 07/30/2021. The results showed that the best model is Ensemble ARMA-GARCH with MSE 1.3231×10-6. At confidence level of 95% and 1 day holding period, the VaR of the Ensemble ARMA-GARCH was -0.0213. Based on the Backtesting test, it is proven to be very accurate to predict the value of loss risk because the value of the Violation Ratio (VR) is equal to 0.

MODELING OF SEA SURFACE TEMPERATURE BASED ON PARTIAL LEAST SQUARE - STRUCTURAL EQUATION

Variability of Sea Surface Temperature (SST) is one of the climatic features that influence global and regional climate dynamics. Missing data (gaps) in the SST dataset are worth investigating since they may statistically alter the value of the SST change. The partial least square-structural equation modeling (PLS-SEM) approach is used in this work to estimate the causality relationships between exogenous and endogenous latent variables. The findings of this study, which are significant indicators that have a loading factor value > 0.7 are as follows: i) sea surface temperature (oC) as a measure of the latent variable changes in SST, ii) wind speed (m/s) and relative humidity (%) as a measure of the latent variable of weather, and iii) air temperature (oC), long-wave solar radiation (w/m2) as a measure of climate latent variables. The size of the Rsquare value is influenced by the number of gaps. The results of the boostrapping show that the latent variables of weather and climate have a significant effect on changes in SST which are indicated by the value of tstatistics > ttabel. The structural model obtained Changes in SST (η) = -0.330 weather + 0.793 climate + ζ. The model shows that the weather has a negative coefficient, which means that the better the weather conditions, the lower the SST changes. Climate has a positive coefficient, which means that the better the climate, the SST changes will also increase. Rising sea surface temperatures caused by an increase in climate can lead to global warming, impacting El-Nino and La-Nina events.

THE APPLICATION OF THE SEMIPARAMETRIC GSTAR MODEL IN DETERMINING GAMMA-RAY LOG DATA ON SOIL LAYERS

This research examines the semiparametric Generalized Space-Time Autoregressive (GSTAR) spacetime modeling and determines its spatial weight. In general, the spatial weights used are uniform, binary weights, and based on the distance, the result is a fixed weight. The GSTAR model is a stochastic model that takes into account its random variables. Thus, it is necessary to study the random spatial weights. This study introduced a new method to estimate the observed value of the GSTAR model semiparametric with a uniform kernel. The data involved the Gamma Ray (GR) log data on four coal drill holes. The semiparametric GSTAR modeling aimed to predict the amount of log GR in the unobserved soil layer based on the observation data information on the layer above it and its surrounding location. The results revealed that semiparametric GSTAR modeling could predict the presence of coal seams and their thickness of drill holes. The results also highlight the validity test on the out-sample data that the error in each borehole results in a small error. In addition, the error tends to approach the actual observed value at a depth of 1 meter down.

PREDICTION OF FOREST FIRE USING NEURAL NETWORKS WITH BACKPROPAGATION LEARNING AND EXREME LEARNING MACHINE APPROACH USING METEOROLOGICAL AND WEATHER INDEX VARIABLES

Forest fire is one of important catastrophic events and have great impact on environment, infrastructure and human life. In this study, we discuss the method for prediction of the size of the forest fire using the hybrid approach between Fuzzy-C-Means clustering (FCM) and Neural Networks (NN) classification with backpropagation learning and extreme learning machine approach. For comparison purpose, we consider a similar hybrid approach, i.e., FCM with the classical Support Vector Machine (SVM) classification approach. In the empirical study, we apply the considered methods using several meteorological and Forest Weather Index (FWI) variables. We found that the best approach will be obtained using hybrid FCM-SVM for data training, where the best performance obtains for hybrid FCM-NN-backpropagation for data testing.

MODELING OF LOCAL POLYNOMIAL KERNEL NONPARAMETRIC REGRESSION FOR COVID DAILY CASES IN SEMARANG CITY, INDONESIA

Coronavirus disease 2019 (COVID-19) is an infectious disease caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) which was recently discovered. Coronavirus disease is now a pandemic that occurs in many countries in the world, one of which is Indonesia. One of the cities in Indonesia that has found many COVID cases is Semarang city, located in Central Java. Data on cases of COVID patients in Semarang City which are measured daily do not form a certain distribution pattern. We can build a model with a flexible statistical approach without any assumptions that must be used, namely the nonparametric regression. The nonparametric regression in this research using Local Polynomial Kernel approach. Determination of the polynomial order and optimal bandwidth in Local Polynomial Kernel Regression modeling use the GCV (Generalized Cross Validation) method. The data used this research are data on the number of COVID patients daily cases in Semarang, Indonesia. Based on the results of the application of the COVID patient daily cases in Semarang City, the optimal bandwidth value is 0.86 and the polynomial order is 4 with the minimum GCV is 3179.568 so that the model estimation results the MSE is 2922.22 and the determination coefficient is 97%. The estimation results show the highest number of Corona in the Semarang City at the beginning of July 2020. After the corona case increased in July, while the corona case in August decreased.

VARIANCE GAMMA PROCESS WITH MONTE CARLO SIMULATION AND CLOSED FORM APPROACH FOR EUROPEAN CALL OPTION PRICE DETERMINATION

The Option is widely applied in the financial sector. The Black-Scholes-Merton model is often used in calculating option prices on a stock price movement. The model uses geometric Brownian motion which assumes that the data is normally distributed. However, in reality, stock price movements can cause sharp spikes in data, resulting in nonnormal data distribution. So we need a stock price model that is not normally distributed. One of the fastest growing stock price models today is the process exponential model. The process has the ability to model data that has excess kurtosis and a longer tail (heavy tail) compared to the normal distribution. One of the members of the process is the Variance Gamma (VG) process. The VG process has three parameters which each of them, to control volatility, kurtosis and skewness. In this research, the secondary data samples of options and stocks of two companies were used, namely zoom video communications, Inc. (ZM) and Nokia Corporation (NOK). The price of call options is determined by using closed form equations and Monte Carlo simulation. The Simulation was carried out for various values until convergent result was obtained.

RELATIVE RISK OF CORONAVIRUS DISEASE (COVID-19) IN SOUTH SULAWESI PROVINCE, INDONESIA: BAYESIAN SPATIAL MODELING

The Covid-19 has exploded in the world since late 2019. South Sulawesi Province has the highest number of Covid-19 cases outside Java Island in Indonesia. This paper aims to determine the most suitable Bayesian spatial conditional autoregressive (CAR) localised models in modeling the relative risk (RR) of Covid-19 in South Sulawesi Province, Indonesia. Bayesian spatial CAR localised models with different hyperpriors were performed adopting a Poisson distribution for the confirmed Covid-19 counts to examine the grouping of Covid-19 cases. All confirmed cases of Covid-19 (19 March 2020-18 February 2021) for each district were included. Overall, Bayesian CAR localised model with G = 5 with a hyperprior IG (1, 0.1) is the preferred model to estimate the RR based on the two criteria used. Makassar and Toraja Utara have the highest and the lowest RR, respectively. The group formed in the localised model is influenced by the magnitude of the mean and variance in the count data between areas. Using suitable Bayesian spatial CAR localised models enables the identification of high-risk areas of Covid-19 cases. This localised model could be applied in other case studies.

COMPARATIVE STUDY OF DISTANCE MEASURES ON FUZZY SUBTRACTIVE CLUSTERING

Clustering is a data analysis process which applied to classify the unlabeled data. Fuzzy clustering is a clustering method based on membership value which enclosing set of fuzzy as a measurement base for classification process. Fuzzy Subtractive Clustering (FSC) is included in one of fuzzy clustering method. This research applies Hamming distance and combined Minkowski Chebysev distance as a distance parameter in Fuzzy Subtractive Clustering. The objective of this research is to compare the output quality of the cluster from Fuzzy Subtractive Clustering by using Hamming distance and combine Minkowski Chebysev distance. The comparison of the two distances aims to see how well the clusters are produced from two different distances. The data used is data on hypertension. The variables used are age, gender, systolic pressure, diastolic pressure, and body weight. This research shows that the Partition Coefficient value resulted on Fuzzy Subtractive Clustering by applying combined Minkowski Chebysev distance is higher than the application of Hamming distance. Based on this, it can be concluded that in this study the quality of the cluster output using the combined Minkowski Chebysev distance is better.

A STUDY OF GENERALIZED LINEAR MIXED MODEL FOR COUNT DATA USING HIERARCHICAL BAYES METHOD

Poisson Log-Normal Model is one of the hierarchical mixed models that can be used for count data. Several estimation methods can be used to estimate the model parameters. The first objective of this study was to examine the performance of the parameter estimator and model built using the Hierarchical Bayes method via Markov Chain Monte Carlo (MCMC) with simulation. The second objective was applied the Poisson Log-Normal model to the West Java illiteracy Cases data which is sourced from the Susenas data on March 2019. In 2019, the incidence of illiteracy is a very rare occurrence in West Java Province. So that, it is suitable as an application case in this study. The simulation results showed that the Hierarchical Bayes parameter estimator through MCMC has the smallest Root Mean Squared Error of Prediction (RMSEP) value and the absolute bias is relatively mostly similar when compared to the Maximum Likelihood (ML) and Penalized Quasi-Likelihood (PQL) methods. Meanwhile, the empirical results showed that the fixed variable is the number of respondents who have a maximum education of elementary school have the greatest risk of illiteracy. Also, the diversity of census blocks significantly affects illiteracy cases in West Java 2019.

AUTOREGRESSIVE FRACTIONAL INTEGRATED MOVING AVERAGE (ARFIMA) MODEL TO PREDICT COVID-19 PANDEMIC CASES IN INDONESIA

Currently the emergence of the novel coronavirus (Sars-Cov-2), which causes the COVID-19 pandemic and has become a serious health problem because of the high risk causes of death. Therefore, fast and appropriate action is needed to reduce the spread of the COVID-19 pandemic. One of the way is to build a prediction model so that it can be a reference in taking steps to overcome them. Because of the nature of transmission of this disease which is so fast and massive cause extreme data fluctuations and between objects whose observational distances are far enough correlated with each other (long memory). The result of this determination is the best ARFIMA model obtained to predict additional of recovering cases of COVID-19 is (1,0,489.0) with an SMAPE value of 12,44%, while the case of death is (1.0.429.0) with SMAPE value of 13,52%. This shows that the ARFIMA model can accommodate well the long memory effect, resulting in a small bias. Also in estimating model parameters, it is also simpler. For cases of recovery and death, the number is increasing even though the case of death is still very high compared to cases of recovery.