Performance of mortality rates using deep learning approach / Mohamad Hasif Azim and Saiful Izzuan Hussain
Mortality has a vital role in population dynamics and is critical in a wide variety of fields, including demography, economics, and social sciences. This study aims to model and compare the mortality rate using two different models; the Lee-Carter model and Deep Neural Network (DNN). The sample data...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Book |
| Published: |
2021.
|
| Subjects: | |
| Online Access: | Link Metadata |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
MARC
| LEADER | 00000 am a22000003u 4500 | ||
|---|---|---|---|
| 001 | repouitm_56136 | ||
| 042 | |a dc | ||
| 100 | 1 | 0 | |a Azim, Mohamad Hasif |e author |
| 700 | 1 | 0 | |a Hussain, Saiful Izzuan |e author |
| 245 | 0 | 0 | |a Performance of mortality rates using deep learning approach / Mohamad Hasif Azim and Saiful Izzuan Hussain |
| 260 | |c 2021. | ||
| 500 | |a https://ir.uitm.edu.my/id/eprint/56136/1/56136.pdf | ||
| 520 | |a Mortality has a vital role in population dynamics and is critical in a wide variety of fields, including demography, economics, and social sciences. This study aims to model and compare the mortality rate using two different models; the Lee-Carter model and Deep Neural Network (DNN). The sample data used is the case of the United Kingdom population. Mortality rates were modeled with the Lee-Carter model and deviance goodness of fit were used to test the model's suitability of the data. Next, mortality rates are modeled with the Deep Neural Network (DNN) and both models are compared based on the mean square error (MSE) values. The results showed that the DNN model fits the best. Overall, we conclude that DNN approach appears to be a potential model to model and forecast population mortality. | ||
| 546 | |a en | ||
| 690 | |a HA Statistics | ||
| 690 | |a Statistical data | ||
| 655 | 7 | |a Conference or Workshop Item |2 local | |
| 655 | 7 | |a PeerReviewed |2 local | |
| 787 | 0 | |n https://ir.uitm.edu.my/id/eprint/56136/ | |
| 856 | 4 | 1 | |u https://ir.uitm.edu.my/id/eprint/56136/ |z Link Metadata |