A multistage risk-averse stochastic programming model for personal savings accrual: the evidence from Lithuania

Abstract

In this paper we consider the problem of choosing the optimal pension fund in the second pillar of Lithuanian pension system by providing some guidelines to individuals with defined contribution pension plans. A multistage risk-averse stochastic optimization model is proposed that can be used to plan a long-term pension accrual under two different cases: minimum and maximum accumulation plans as possible options in the system. The investment strategy of personal savings is based on the optimal solutions over possible scenario realizations generated for a particular participant. The concept of the risk-averse decision-maker is implemented by choosing the Conditional Value at Risk (CVaR) as the risk measure defined by a nested formulation that guarantees the time consistency in the multistage model. The paper focuses on three important decision-making moments corresponding to the duration of periods to be modelled. The first period is a short-term accumulation, while the second period is a long-term accumulation with possibly high deviation of objective function value. The third period is designed to implement the concept of Target Date Fund in the second pillar pension scheme as the subsequent need to protect against potential losses at risky pension funds. The experimental findings of this research provide insights for individuals as decision-makers to select pension funds, as well as for policy-makers by revealing the vulnerability of pension system.

Dominance-Based Decision Rules for Pension Fund Selection under Different Distributional Assumptions

“…In this case, only funds with R = 1 could be used as base assets. Finally, dynamic (multi-stage) portfolio (or asset-liability) management models [5,8,27] could be extended in order to find the optimal investment strategy for a participant. These models could either maximise the dominance ratio or at least keep it reasonably high (only portfolios with sufficiently high dominance ratios would be feasible).…”

“…There are several reasons the choice of a pension fund is complicated. First, due to a lack of financial literacy and understanding, a participant might select an inappropriate fund for their pension savings [4,5]. In response to this problem, pension fund managers usually report simple descriptive statistics, accompanied by some explanations, related to the historical performances of their pension funds, but no deeper analyses or forecasts are published.…”

“…In particular, empirical, student-t, hyperbolic and normal-inverse Gaussian (NIG) distributions are used to describe the PFs' returns. In fact, in the last few decades, many papers have published evidence of the non-normality (see [30][31][32]) and fatness (see [5,33,34]) of financial time series. Furthermore, some researchers have found evidence that the tails of returns are typically semi-heavy (see [35][36][37]) from a class of generalised hyperbolic distributions, which includes the student-t, hyperbolic and NIG distributions.…”

Clustering of Latvian Pension Funds Using Convolutional Neural Network Extracted Features

“…The second and third pillars are based on the Anglo-Saxon model with state-funded and supplementary/voluntary schemes. Furthermore, the second pillar is based on defined contribution (DC) plans managed by private companies (for more details about the Latvian system, see [12] or [13] about Lithuania). While the second pillar is mainly compulsory in Latvia, the third pillar is entirely voluntary.…”

“…Figure13. The frequencies of different cluster combinations using other clustering algorithms with omitted OPTICS results clustered the dataset into two clusters with silhouette scores above the mean score.…”