Συγκρίσεις εναλλακτικών μεθόδων παλινδρόμησης σε οικονομετρική ανάλυση μικροδεδομένων : γραμμική και παλινδρόμηση πεμπτημορίων

Καλογερά, Κωνσταντίνα

Master Thesis

Author

Καλογερά, Κωνσταντίνα

Date

2023-11

Abstract

Multiple Linear Regression is the most common method of detecting the conditional mean behavior of a variable, when is affected by a set of different variables. On the contrary, Quantile Regression studies the behavior of the response variable over the entire range of its distribution. The current study compares the models, their particularities and their differences. Furthermore, data from the 8 th wave of SHARE survey were used to detecting which factors affect the annual pension amount. The statistical analysis of the data was realized using SPSS 26, using regression models. The purpose of the study was to compare the two regression models to determine which had better interpretive capacity at the height of pensions. From the analysis, Linear Regression assumptions were not met for the conclusions to be reliable, while Quantile Regression could possibly separate the low from the high levels of the pension amount.

Postgraduate Studies Programme

Εφαρμοσμένη Στατιστική

Department

Σχολή Χρηματοοικονομικής και Στατιστικής. Τμήμα Στατιστικής και Ασφαλιστικής Επιστήμης

Number of pages

Language

Greek

URI

https://dione.lib.unipi.gr/xmlui/handle/unipi/16085
http://dx.doi.org/10.26267/unipi_dione/3507

Collections

Τμήμα Στατιστικής και Ασφαλιστικής Επιστήμης

Show full item record

Except where otherwise noted, this item's license is described as
Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα