Modeling Stated Severance-Pay Withdrawal Intentions in a Mandatory-Pension System: Interpretable Statistical Learning for Computational Actuarial Modeling

Pre-retirement withdrawal of pension-related severance capital may weaken retirement income adequacy. This study examines stated withdrawal intentions in Israel’s mandatory-pension system using a survey conducted from December 2023 to May 2024. The complete-case analytic sample comprises 421 salaried respondents with observed values for the four-category outcome and all 17 predictors. Seven classifiers were evaluated using stratified five-fold cross-validation repeated three times. Gradient boosting achieved the highest mean AUC (0.821), compared with 0.588 for additive logistic regression. A separate nested five-fold evaluation repeated four times, with tuning confined to inner folds, yielded a mean outer-fold AUC of 0.851 (SD 0.045). An additional spline-logistic benchmark, ordinal and multinomial regressions, and cross-fitted SHAP analyses examined model specification and interpretability. Pension centrality and impatience were the leading global SHAP features. These findings support nonlinear prediction of stated intentions, without implying observed withdrawals or causal effects.

Authors

Institutions

Publication Details

Journal
Mathematics
Published
2026-11-13
DOI
https://doi.org/10.3390/math14224169
Primary Topic
Financial Literacy, Pension, Retirement Analysis
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Modeling Stated Severance-Pay Withdrawal Intentions in a Mandatory-Pension System: Interpretable Statistical Learning for Computational Actuarial Modeling

Shenhav Malul
1 links
Mathematics
Financial Literacy, Pension, Retirement Analysis
article

Modeling Stated Severance-Pay Withdrawal Intentions in a Mandatory-Pension System: Interpretable Statistical Learning for Computational Actuarial Modeling

Shenhav Malul
article en

Abstract

Pre-retirement withdrawal of pension-related severance capital may weaken retirement income adequacy. This study examines stated withdrawal intentions in Israel’s mandatory-pension system using a survey conducted from December 2023 to May 2024. The complete-case analytic sample comprises 421 salaried respondents with observed values for the four-category outcome and all 17 predictors. Seven classifiers were evaluated using stratified five-fold cross-validation repeated three times. Gradient boosting achieved the highest mean AUC (0.821), compared with 0.588 for additive logistic regression. A separate nested five-fold evaluation repeated four times, with tuning confined to inner folds, yielded a mean outer-fold AUC of 0.851 (SD 0.045). An additional spline-logistic benchmark, ordinal and multinomial regressions, and cross-fitted SHAP analyses examined model specification and interpretability. Pension centrality and impatience were the leading global SHAP features. These findings support nonlinear prediction of stated intentions, without implying observed withdrawals or causal effects.

MathematicsVol. 14(22)
Ben-Gurion University of the Negev (IL)
Financial Literacy, Pension, Retirement Analysis
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon