A model-based framework for probabilistic simulation of legal policies

Abstract

Abstract-Legal policy simulation is an important decisionsupport tool in domains such as taxation. The primary goal of legal policy simulation is predicting how changes in the law affect measures of interest, e.g., revenue. Currently, legal policies are simulated via a combination of spreadsheets and software code. This poses a validation challenge both due to complexity reasons and due to legal experts lacking the expertise to understand software code. A further challenge is that representative data for simulation may be unavailable, thus necessitating a data generator. We develop a framework for legal policy simulation that is aimed at addressing these challenges. The framework uses models for specifying both legal policies and the probabilistic characteristics of the underlying population. We devise an automated algorithm for simulation data generation. We evaluate our framework through a case study on Luxembourg's Tax Law.

Model-Based Simulation of Legal Requirements: Experience from Tax Policy Simulation

“…The OCL expressions for the inputs are not shown in the policy model of Fig. 2 (see [8], [9] for examples of policy models with their inputs expressed in OCL).…”

“…An update function is an action performed within the workflow of a policy in order to record the outcome resulting from the simulation of that policy (see [9] for a complete list of the stereotypes in our customized activity diagram notation).…”

“…The goal of simulation is to predict the impact of a set of proposed changes to laws and regulations, and ensure that the proposed changes will bring about the desired outcomes without unwanted side effects. This new application of policy models prompted us to develop a modelbased simulator [9], drawing on the same modeling approach developed previously for legal compliance testing.…”

“…While the feasibility of building individual policy models was examined in our prior work [8], [9], we did not previously investigate whether one can build a seamless set of models that need to be considered together in a real-world simulation scenario. Furthermore, our simulator includes a probabilistic data generator to create artificial data in situations where access to real simulation data is restricted (the situation we have to deal with in our case study), or where real data is unavailable, e.g., when we are simulating a new policy for which no historical data exists.…”

“…We therefore address only the inputs to and the outputs from our simulation approach, rather than the technical machinery behind the approach. A detailed treatment of the technical components of the approach is available in our previous work [9]. Below, we present the modeling aspects of our approach and outline tool support.…”