Search for: All records

Reliable GDP forecasting plays a critical role in macroeconomic management, particularly during periods of heightened uncertainty such as the 2008 financial crisis and the 2020 pandemic shock. During these episodes, timely GDP projections provided policymakers with essential guidance for assessing recession depth, calibrating emergency fiscal and monetary responses, and preventing further macroeconomic deterioration. To support policy evaluation and medium-term growth planning, this study applies a unified forecasting framework to annual U.S. GDP data from 1980 to 2024 using three representative base models—ARIMA, OLS, and LSTM—together with a linear hybrid ensemble that combines their predictions. Each base model is paired with preprocessing tailored to its statistical assumptions. For ARIMA, unit-root testing confirms non-stationarity, and second-order differencing produces a stationary series, leading to the selection of an ARIMA (0,2,2) specification via information criteria and autocorrelation diagnostics. For OLS, a log-linear specification stabilizes exponential growth, yielding an estimated long-run annual trend of approximately 4.98%. For LSTM, the log-transformed series is normalized and converted into four-year sliding window sequences to enable nonlinear temporal learning despite the limited sample size. The hybrid ensemble—implemented as a convex linear blend of the three base forecasts estimated by least squares—is designed to exploit complementary strengths across linear and nonlinear components. All models generate forecasts for 2025–2030 and are evaluated using MAE, RMSE, MAPE, MASE, and Theil’s U. The empirical results show that the hybrid ensemble (COLBA) model achieves the lowest MAE, RMSE, MASE, and Theil’s U, while the standalone ARIMA model attains the smallest MAPE. These findings indicate that combining statistically grounded econometric models with neural networks can modestly improve overall accuracy even when only annual data are available, while ARIMA remains particularly effective for preserving proportional accuracy in trend-dominated macroeconomic series. 

To address the growing challenges posed by Cyber threats, anti-malware organizations have increasingly turned to Machine Learning (ML). In recent years, machine learning algorithms have become indispensable for solving complex classification problems, outperforming traditional statistical methods by capturing intricate patterns in high dimensional data. However, selecting the optimal model requires rigorous evaluation in multiple performance metrics while ensuring stability across different data splits. In this study, we conducted a comprehensive assessment of eight machine learning algorithms. Random Forest (RF), Extreme Gradient Boosting (XGBoost), Support Vector Machine (SVM), Logistic Regression (LR), Naive Bayes, Light Gradient Boosting Machine (LightGBM), Decision Tree (DT), and k-Nearest Neighbors (KNN) using stratified 5-fold cross-validation. Our results reveal that RF, LightGBM, DT, and KNN achieve exceptional performance, with identical near-perfect scores in accuracy (0.9918), precision (0.9920), recall (0.9918), F1 score (0.9918) and Area Under the Receiver Operation Characteristic Curve (AUC-ROC) (0.9998), along with remarkably low variance (10−6 to 10−8), demonstrating unparalleled robustness. The study highlights the superiority of tree-based ensembles and KNN in achieving high predictive power and stability, whereas classical algorithms such as logistic regression and naive Bayes lag. Despite XGBoost’s reputation, its performance here is eclipsed by simpler tree-based methods. Our analysis underscores the importance of considering variance when evaluating model selection, particularly for critical applications where stability is paramount, and provides actionable insights for practitioners seeking reliable, high-accuracy classifiers. 

To address the growing challenges posed by Cyber threats, anti-malware organizations have increasingly turned to Machine Learning (ML). In recent years, machine learning algorithms have become indispensable for solving complex classification problems, outperforming traditional statistical methods by capturing intricate patterns in high dimensional data. However, selecting the optimal model requires rigorous evaluation in multiple performance metrics while ensuring stability across different data splits. In this study, we conducted a comprehensive assessment of eight machine learning algorithms. Random Forest (RF), Extreme Gradient Boosting (XGBoost), Support Vector Machine (SVM), Logistic Regression (LR), Naive Bayes, Light Gradient Boosting Machine (LightGBM), Decision Tree (DT), and k-Nearest Neighbors (KNN) using stratified 5-fold cross-validation. Our results reveal that RF, LightGBM, DT, and KNN achieve exceptional performance, with identical near-perfect scores in accuracy (0.9918), precision (0.9920), recall (0.9918), F1 score (0.9918) and Area Under the Receiver Operation Characteristic Curve (AUC-ROC) (0.9998), along with remarkably low variance (10−6 to 10−8), demonstrating unparalleled robustness. The study highlights the superiority of tree-based ensembles and KNN in achieving high predictive power and stability, whereas classical algorithms such as logistic regression and naive Bayes lag. Despite XGBoost’s reputation, its performance here is eclipsed by simpler tree-based methods. Our analysis underscores the importance of considering variance when evaluating model selection, particularly for critical applications where stability is paramount, and provides actionable insights for practitioners seeking reliable, high-accuracy classifiers.