Coronavirus Disease 2019 (COVID-19) is a new coronavirus that was discovered in Wuhan, China, at the end of 2019. In March 2020, the outbreak extended throughout the world, including Indonesia and one of its provinces, South Kalimantan. This rapid expansion should belinked to people’s mobility between regions, hence the linkage across regions must be examined. In South Kalimantan Province, the purpose of this research is to evaluate the distribution and relationship across regions in terms of the number of positive COVID-19 cases, the number of additional positive COVID-19 cases, and the number of COVID-19 patients under treatment. The spatial autocorrelation analysis with the Moran Index and Local Indicator of Spatial Autocorrelation (LISA) tests were used to determine the spatial autocorrelation between what and what using what data/where the data obtained? from March 22 to September 30, 2020. Based on the results of the Moran Index test, it is known that there is a spatial autocorrelation in the number of cases, the number of additional cases and the number of positive confirmed COVID-19 patients in treatment between one region and another neighboring area. While the results of the LISA Index test show that Balangan Regency, Hulu Sungai Tengah Regency, Hulu Sungai Utara Regency, Banjarmasin City, Tabalong Regency and Banjar Regency affect the level of COVID-19 cases in their respective neighboring areas. Therefore, there is a need for policies to control community mobility in those spatially correlated areas and increase testing and tracing to control the spread of COVID-19 cases in South Kalimantan Province.

Bad credit card is a problem of inability of credit card users to pay credit card bills that can cause losses to both parties concerned. In order to avoid losses caused by bad credit cards, the provider must conduct a careful analysis of prospective or old customers using credit cards. This study aims to classify bad credit card customers using machine learning techniques, namely classification techniques. One of the classification techniques used is the XGBoost method which is useful for regression analysis and classification based on the Gradient Boosting Decision Tree (GBDT), the XGBoost method has several hyperparameters that can be configured to improve the performance of the model. Hyperparameter tuning method used is grid search cross validation which is then validated using 10-Fold Cross Validation. XGBoost hyperparameters configured include n_estimators, max_depth, subsample, gamma, colsample_bylevel, min_child_weight and learning_rate. Based on the results of this study proves that the use of algorithms with hyperparameter tuning can improve the performance of eXtreme Gradient Boosting algorithm in the process of classification of credit card customers with an accuracy of 80.039%, precision of 81.338% and a recall value of 96.854%.   Keywords: XGBoost, classification, Accuracy, Precision, Recall

Salah satu aspek penting dalam perekonomian suatu negara adalah tingkat pendapatan yang biasanya berasal dari proses transaksi di dalam negara tersebut. Aspek ini sering dipakai sebagai indikator untuk melihat laju pertumbuhan ekonomi baik dari tingkatan daerah maupun nasional. Pada tingkat daerah, indikator yang dimaksud disebut sebagai Produk Domestik Regional Bruto (PDRB). Ada banyak variabel yang menyumbang struktur PDRB dan bisa digambarkan dalam bentuk model eksplanatoris. Dalam mengembangkan model eksplanatoris yang dapat menggambarkan pola hubungan antara variabel respon dan dua/lebih variabel bebas ini dapat dilakukan melalui pendekatan analisis regresi. Namun, model yang terbentuk dengan pendekatan regresi masih bersifat global karena diberlakukan pada seluruh lokasi pengamatan. Dengan kata lain, pendekatan model global ini biasanya menggunakan rata-rata dari wilayah lokal (area yang lebih kecil luasannya pada wilayah teregionalisasi). Jika tidak ada atau hanya sedikit keragaman antar wilayah lokal, maka pendekatan model regresi secara global akan memberikan informasi yang akurat. Oleh karena itu, apabila kondisi pengamatan di lokasi yang satu dengan lokasi yang tidak selalu sama karena dipengaruhi oleh efek spasial (lokasi) maka model regresi dapat dibentuk dengan menambahkan efek spasial yang sering dikenal sebagai model Geographically Weigthed Regression (GWR). Dalam penelitian ini dilakukan analisa pembentukan model regresi yang tepat untuk kasus PDRB di Provinsi Kalimantan Selatan.

Insurance is an attempt of risk diversion by the insured person to the insurance company. The risk is referred to the future event that will potentially cause a financial loss. Based on many risk factors,the status of insurance was divided into a single decrement and a multiple decrement. In single decrement, the only factor caused benefit payment is death, while in multiple decrement there is more than one factors caused benefit payment. As a consequence, beside the random variable of time until termination , there is another random variable appears that is the cause of decrement . The aim of this study was to describe the development process of a multiple decrement table and determine net single premium based on multiple decrement status. This study was conducted by describing the construction process of components in the multiple decrement table using joint distribution and marginal distribution for each random variable. This study is a various equation for constructing a multiple decrement table was obtained. That probability equation was also used to form the net single premium equation of term insurance based on multiple decrement status by using probability function of time until termination and cause of termination. Keywords: Term Insurance, Multiple Decrement, Net Single  Premium

In everyday life, fuel oil is quite important. The need for fuel is increasing every day, which means that the supply of fuel oil must keep up with the demand. As a reason, we require a way for predicting future fuel needs. The forecasting method is one that is frequently utilized. Forecasting is a method for predicting future conditions based on historical data. The transfer function approach is one way to forecast data with several variables in time series analysis. The objective of this research is to estimate the parameters of the transfer function model and use a transfer function approach to predict the movement of fuel, particularly pertalite. The parameter estimation results in this research are ω ^ 0 = 0.033 ; ω ^ 1 = − 0.0358 ; ω ^ 2 = 0.0627 ; δ ^ 1 = − 0.9713 ; δ ^ 2 = 1 ; θ ^ 1 = − 0.9141 , and the forecast value for the 214th period is 8762.61, based on the data used, namely for 213 periods starting from the 1st period until the 213th period.