Institutional Repository of Vilnius University
Mirtingumo analizė naudojant Lee-Carter modelį su rotacija /
Abstract
dc:descriptionIn this paper, the Lee-Carter model and its modification with rotation are analysed and applied to mortality forecasting. A data analysis of four countries was conducted: Lithuania, Japan, the United States and Switzerland. Forecasting accuracy is evaluated using the Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE) metrics, which allow distinguishing the overall forecast deviation from large local errors in specific age groups. The results show that the benefit of rotation appears in countries where mortality is low and steadily declining. In Japan, the values of both metrics decreased significantly after applying rotation. In the United States a clear improvement in both metrics is also observed. In Switzerland, rotation proves more effective at reducing large local errors (RMSE) than overall deviation (MAE). In Lithuania, the benefit of rotation is minimal, as the volatility of mortality due to post-Soviet social changes leads to errors that rotation cannot correct. In conclusion, the Lee-Carter model with rotation is a meaningful improvement over the standard model, but its effectiveness depends on the nature of a country's mortality history.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Jankauskas, Mantas,
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD308326985&prefLang=en_US