Article Per Year (5 Year)
p-Index From 2021 - 2026
0.408
P-Index
This Author published in this journals
All Journal bit-Tech
Kartika Maulida Hindrayani
Universitas Pembangunan Nasional Veteran Jawa Timur
Computer Science & IT

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
ARIMA-TGARCH Model for Return Prediction and Risk Estimation with VaR Imanta Ginting; Trimono Trimono; Kartika Maulida Hindrayani
bit-Tech Vol. 8 No. 2 (2025): bit-Tech
Publisher : Komunitas Dosen Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32877/bt.v8i2.3090

Abstract

Investment activity in the Indonesian capital market has experienced significant growth, driven by increasing public awareness and accessibility to financial instruments. Stocks remain the most favored investment tool due to their potential for high returns, though they come with higher risks. Accurate modeling of return dynamics and risk estimation is thus crucial for informed investment decisions. This study analyzes the return and volatility of PT Telekomunikasi Indonesia Tbk (TLKM) stock using a hybrid time series approach that combines the Autoregressive Integrated Moving Average (ARIMA) model and the Threshold Generalized Autoregressive Conditional Heteroskedasticity (TGARCH) model. The analysis uses daily closing price data from 2020 to 2024, with 1,210 observations. The best-fitting model, ARIMA(2,0,2)–TGARCH(1,1), resulted in low Root Mean Squared Error (RMSE) values of 0.0188 for both training and testing datasets, indicating strong prediction accuracy. Forecasting over a five-day horizon revealed fluctuating returns and a decreasing trend in volatility, from 0.0230 to 0.0198. Additionally, the study utilized the Value at Risk (VaR) method to estimate potential losses under normal market conditions. At a 95% confidence level, the predicted daily loss for a capital investment of IDR 50,000,000 ranged between IDR 1,633,108 and IDR 1,859,355. The combination of ARIMA and TGARCH, integrated with VaR, provides a comprehensive framework for capturing both linear return trends and asymmetric volatility, offering investors a robust quantitative tool for managing risks and optimizing strategies.
Heckman Probit Two-Step Regression Approach for Analyzing Open Unemployment Factors in West Java Province Holly Patrycia; Dwi Arman Prasetya; Kartika Maulida Hindrayani
bit-Tech Vol. 8 No. 2 (2025): bit-Tech
Publisher : Komunitas Dosen Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32877/bt.v8i2.3096

Abstract

Open unemployment remains a major socio-economic challenge in Indonesia, with West Java recording the highest national rate in August 2024 at 6.75%. This study investigates the determinants of open unemployment using the Heckman Probit Two-Step model, an approach rarely applied in Indonesian labor market research. Unlike conventional regression methods, this model corrects for sample selection bias by simultaneously estimating labor force participation and unemployment status. Data are drawn from the 2024 Survei Angkatan Kerja Nasional (SAKERNAS) conducted by Badan Pusat Statistik (BPS), covering working-age individuals in West Java Province. The first stage models labor force entry, while the second stage incorporates the Inverse Mills Ratio (IMR) to adjust for selection effects. Results show that the IMR coefficient (–0.3100, p = 0.0412) is statistically significant, confirming the necessity of the two-step correction. The explanatory power of the model is substantial, with Pseudo-R² values of 0.385 for labor force participation and 0.381 for open unemployment. Marginal effects indicate that being married reduces unemployment probability by 5.50%, each additional year of age decreases it by 2.79%, whereas a longer job search increases it by 3.35%. Training experience lowers unemployment risk, while disabilities and larger household size increase vulnerability. Methodologically, the study demonstrates the advantages of Heckprobit in producing unbiased estimates compared to descriptive or conventional probit approaches previously used in Indonesia. Nonetheless, the cross-sectional design and focus on a single province limit generalizability. Findings provide valuable evidence for policymakers to design targeted, inclusive employment strategies aligned with regional development goals
Co-Authors Dwi Arman Prasetya Holly Patrycia Imanta Ginting Trimono, Trimono