Dynamic Programming and Optimal Growth Models

Dynamic programming is a methodological framework for solving optimisation problems that evolve over time by breaking them into simpler subproblems. Central to this approach is the principle of optimality, which asserts that an optimal policy at any point must comprise optimal decisions in subsequent periods. In continuous time, this leads to the Hamilton–Jacobi–Bellman equation, while in discrete settings it gives rise to the Bellman equation. Optimal growth models apply these tools to analyse how economies allocate resources—such as capital, labour and consumption—over time to maximise welfare or output. Classical examples include the Ramsey–Cass–Koopmans model in deterministic form and its stochastic extensions, which introduce uncertainty in productivity or preferences. Recent innovations have addressed non-convex technologies, habit formation, recursive utility functions and intergenerational externalities, enhancing our understanding of macroeconomic stability, policy trade-offs and resource management under uncertainty. Computational advances, such as value-function iteration, policy iteration and grid-based methods, now allow for high-dimensional problems, bringing dynamic programming ever closer to applications in climate policy, finance and ecological modelling.

Research from Nature Portfolio

No recent Nature Portfolio content available.

Dynamic Programming and Optimal Growth Models publication trend

The graph below shows the total number of articles in dynamic programming and optimal growth models across all publications each year (not limited to Nature Index journals).

Technical terms

Bellman equation: A recursive functional equation characterising the value function in discrete-time dynamic optimisation.

Hamilton–Jacobi–Bellman equation: A partial differential equation describing the value function in continuous-time control problems.

Value function: A mapping from a state to the maximum attainable objective value under an optimal policy.

Policy function: A rule that specifies the optimal decision as a function of the current state.

Markov perfect equilibrium: A strategy profile in a dynamic game where each player’s strategy depends only on the current state and is optimal given other players’ strategies.

Endogenous gridpoint method: A numerical technique that constructs the policy function by choosing grid points on the post-decision state rather than on the pre-decision state.

References

1. On recursive utilities with non-affine aggregator and conditional certainty equivalent. Economic Theory (2019).
2. Existence of Stationary Markov Perfect Equilibria in Stochastic Altruistic Growth Economies. Journal of Optimization Theory and Applications (2014).
3. Equilibrium Dynamics in the Neoclassical Growth Model with Habit Formation and Elastic Labor Supply. Theoretical Economics Letters (2012).
4. Optimal Growth and Borrowing with Non-convex Technology. East Asian Economic Review (2006).
5. Optimal Consumption in a Stochastic Ramsey Model with Cobb‐Douglas Production Function. International Journal of Mathematics and Mathematical Sciences (2013).
6. Monotonicity of savings function in Endogenous Gridpoint Method with stochastic portfolio returns. Economics Letters (2024).