Χρήση τεχνικών ανάλυσης μεγάλων δεδομένων στον αναλογισμό
Use of big data analytics in actuary
View/
Keywords
Μεγάλα δεδομένα ; Βig data ; Actuary ; Actuarial ; Random forest ; KNN ; Λογιστική παλινδρόμηση ; Μηχανική μάθησηAbstract
Nowadays there is a need to process huge amounts of high dimensional big data with high complexity. The scientific field of big data analytics has now had a positive impact on the insurance industry, since already in large insurance companies there is an extensive use of such techniques in premium pricing and claims management. For instance, in pricing, the application of segmentation techniques and predictive modeling is particularly beneficial to assess the risk more accurately and to make safe / beneficial insurance premiums. Thus, insurance companies, now, use advanced tools e.g. Behavioral models based on customer profile data - with a continuous flow of real-time data - e.g. satellite data, weather reports, vehicle sensors - to create a detailed and personalized risk assessment.
In this thesis, a systematic presentation of statistical techniques of big data analysis that have been used or developed exclusively for the field of actuarial and risk management is carried out. In particular, the characteristics of Big Data and the techniques that govern them are described, according to the literature. Next, the classification techniques and decision trees are presented. Finally, specific examples of application some of these techniques are presented to demonstrate how they are used and their effectiveness.
Postgraduate Studies Programme
Αναλογιστική Επιστήμη και Διοικητική ΚινδύνουDepartment
Σχολή Χρηματοοικονομικής και Στατιστικής. Τμήμα Στατιστικής και Ασφαλιστικής ΕπιστήμηςNumber of pages
67Language
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