Penelitian ini bertujuan untuk mendapatkan model terbaik untuk klasifikasi data imbalanced, yaitu  rumah tangga sampel Susenas Maret 2017 di Kabupaten Karangasem, ke dalam kategori miskin atau tidak. Metode yang digunakan adalah Regresi Logistik dan Random Forest dimana masing-masing diterapkan skema cross validation (CV), yaitu stratified 5-fold CV, skema under sampling, oversampling dan combine sampling untuk mengatasi masalah data imbalanced serta proses feature selection. Hasil penelitian menunjukkan bahwa penerapan skema under sampling, oversampling dan combine sampling pada model Regresi Logistik memberikan efek meningkatnya rata-rata nilai sensitivity dan turunnya rata-rata nilai akurasi dan specificity. Sedangkan pada model Random Forest, efek tersebut hanya terlihat dari hasil skema under sampling saja. Proses feature selection dapat menurunkan varian nilai akurasi, specificity, sensitivity dan AUC pada model Regresi Logistik dan Random Forest hanya pada skema tertentu. Model terbaik secara keseluruhan adalah model model Regresi Logistik dengan skema combine sampling dan tanpa proses feature selection dengan rata-rata nilai akurasi, specificity, sensitivity dan AUC masing-masing sebesar 78,13%, 79,16%, 64,44% dan 77,77%.

The Consumer Price Index (CPI), stock prices and the rupiah exchange rate to the US dollar are important macroeconomic variables which their movements show the economic performance and can affect the monetary and fiscal policies of Indonesia. This makes forecasting effort of these variables become important for policy planning. While many previous studies only focus on examining the effect among macroeconomic variables, this study uses ARIMA (univariate method), transfer function and VAR (multivariate methods) to measure the forecasting accuracy and also observing the effect between these macroeconomic variables. The results showed that the multivariate methods gave better explanation about the relationship between variables than the simple one. Otherwise, the results of accuracy comparison showed that the multivariate methods did not always yield better forecast than the simple one, and these conditions in line with the results and conclusions of M3 and M4 competition.

As one of the priority sectors in economic development of Indonesia, tourism is expected to be the main key in accelerating economic and social growth, hence reducing poverty. The tourism performance, especially international tourism market, is highly prone to intervention events that can reduce the number of inbound tourists and produce a negative impact on economic development of the destination country. Therefore, anticipating and mitigating various intervention events is necessary to maintain the performance of the tourism sector in Indonesia. This study investigates the magnitude and patterns of impact of several intervention events on the number of international visitor arrivals via the three main ports of entry of Indonesia, i.e. Soekarno-Hatta Airport, Ngurah Rai Airport, and Batam Port. The multi input intervention models were constructed by covering intervention events, i.e. terrorism, disease pandemic, global financial crisis, natural disaster, and government policy, occurring in a relatively long time span, more than two decades, from January 1999 to August 2020. The results show that an intervention event does not always have a significant impact on the number of international visitor arrivals at the three main ports of entry. Generally, all intervention events can lead to a decrease in the number of international visitor arrivals but with different magnitude and pattern, with the biggest and longest impact is caused by COVID-19 pandemic. The direct or non-delayed pattern of impact only appears for terrorism and natural disaster that affect the number of international visitor arrivals via Ngurah Rai Airport.

This study provides an alternative procedure to produce the penalized spline small area estimation (P-Spline SAE) model by plotting each of auxiliary variable and variable interest as a simple nonlinearity identification and performing the iterative procedure: by estimating the model, producing partial residual plots to check model adequacy and also identify the nonlinearity, and testing the significance of spline term using Restricted Likelihood Ratio Test (RLRT). These procedures are applied to estimate the monthly average per-capita expenditure of district level in the Province of Bali, 2014 using direct survey estimates from the National Socio-Economic Survey (Susenas 2014) and auxiliary variables from the administrative record of village data (PODES 2014). The Fay-Herriot (FH) model (M1) as a benchmark and four P-Spline SAE model, i.e. the P-Spline SAE with spline term: x2 and x3 (M2), x3 and x4 (M3), x2, x3 and x4 (M4), and x1, x2, x3 and x4 (M5), are obtained. From RLRT, the spline term of M3, M4, and M5 are statistically significant. According to the parametric bootstrap of mean squared prediction error (MSPE) and coefficient of variation (CV), the M4 and M5 show a significant improvement with the CV values range from about 1%-6% compared to M1 with a range from about 4%-17%, shows that these two models more efficient. The M3 model shows the opposite performance even though has the smallest AIC value. More detail, the MSPE and CV produced by M4 are slightly better than M5 makes the M4 is the best model in this study.

The economic shock as a result of the Covid-19 pandemic resulted in an economic contraction, including East Java. The shift in consumer behavior in utilizing e-commerce to adapt to pandemic constraints is a good catalyst for stimulating household final consumption as the largest shareholder in the East Java economy. This study tries to identify socio-demo graphic and spatial factors that can be optimized for the acceleration of East Java’s eco nomic growth during the disruption from the demand side. The results of the identification of individual behavior in online shopping as well as district/city aggregates indicate several socio-demographic and spatial factors that can be optimized. Mapping districts/cities based on the dimensions of online shopping activities and ICT infrastructure conditions can be a more specific alternative solution to further optimize the potential for increasing household consumption as an effort to accelerate economic growth during the Covid-19  pandemic appropriate with the characteristics of each region.