Μοντελοποίηση και πρόβλεψη θνησιμότητας με τη χρήση μεθόδων αποθεματοποίησης
Modeling and forecasting mortality using reserving methods
View/
Keywords
Πρόβλεψη θνησιμότητας ; Αναλογιστική επιστήμη ; Chain-Ladder ; Lee-Carter ; Cairns-Blake-Dowd ; Τεχνικές αποθεματοποίησηςAbstract
This thesis investigates and compares three approaches to mortality rate forecasting, aiming to improve predictive accuracy in actuarial and insurance applications. Specifically, it examines the classical reserving model Chain-Ladder (CL) alongside the stochastic mortality models Lee-Carter (LC) and Cairns-Blake-Dowd (CBD). The theoretical foundations and assumptions underlying the LC and CBD models are presented, while the CL model is innovatively adapted and applied to mortality forecasting by treating mortality as a delayed reporting phenomenon, analogous to the handling of Incurred But Not Reported (IBNR) claims in general insurance. The empirical analysis, based on data from the Human Mortality Database for six countries (the United States, Japan, the United Kingdom, Greece, France, and Italy) and both genders, evaluates the predictive performance of the models using MAE, RMSE, and MAPE metrics. The results demonstrate the superior predictive accuracy of the CL model. Overall, the study highlights the potential of reserving techniques in mortality modeling, offering new insights for more effective risk management by insurance and pension institutions.
Postgraduate Studies Programme
Αναλογιστική Επιστήμη και Διαχείριση ΚινδύνωνDepartment
Σχολή Χρηματοοικονομικής και Στατιστικής. Τμήμα Στατιστικής και Ασφαλιστικής ΕπιστήμηςNumber of pages
65Language
GreekCollections
Except where otherwise noted, this item's license is described as
Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα
Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα