Πλοήγηση ανά Συγγραφέα "Mitrakou, Theofania"

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Α Β Γ Δ Ε Ζ Η Θ Ι Κ Λ Μ Ν Ξ Ο Π Ρ Σ Τ Υ Φ Χ Ψ Ω

Τώρα δείχνει 1 - 1 από 1

Modeling and predicting U.S. recessions using threshold - tree structured - logit models
(2025-11-18) Mitrakou, Theofania; Μητράκου, Θεοφανία; Besbeas, Panagiotis; Chasiotis, Vasileios; Vrontos, Ioannis
Forecasting economic recessions is a key component of macroeconomic analysis and policy planning. This thesis focuses on modeling and predicting U.S. economic recessions using binary regression models and modern machine learning techniques. The analysis begins with the application of classical Logit and Probit models and is extended to more advanced and flexible methods such as LASSO, Elastic Net, Random Forests, and Boosting. Emphasis is placed on capturing non-linear relationships through tree-based and threshold models. The empirical study is based on a wide range of macroeconomic and financial indicators, which are incorporated in lagged form to predict the probability of recession at various forecasting horizons: 1, 3, 6, and 12 months. Model performance is evaluated using standard statistical metrics such as accuracy, sensitivity (recall), and the ROC curve. Finally, the aim of the thesis is to identify the most effective predictive models and to better understand which economic indicators are most associated with the onset of recessions.