Mathematical programming (MP) is a widespread approach to depict production and investment decisions of agents in agent-based models (ABM) related to agriculture. However, introducing dynamics and indivisibilities in MP models renders their solution computing time intensive. We present a...

A new algorithm for calibrating agent-based models is proposed, which employs a popular gradient boosting framework. Machine learning techniques are not used to develop a surrogate model, but rather assist in narrowing down the parameter space during the search for optimal parameters. Our...

The computational time required to solve and estimate dynamic economic models is one of the main constraints in empirical research. The Endogenous Grid Method (EGM) proposed by Carroll (2006) is known to offer impressive speed gains over more traditional stochastic dynamic programming methods,...

Non-stationary income processes are standard in quantitative life-cycle models, prompted by the observation that within-cohort income inequality increases with age. This paper generalizes Tauchen (1986) and Rouwenhorst's (1995) discretization methods to non-stationary AR(1) processes. We...

In general, the properties of the conditional distribution of multiple period returns do not follow easily from the one-period data generating process. This renders computation of Value-at-Risk and Expected Shortfall for multiple period returns a non-trivial task. In this paper we consider some...

We propose a simple risk-adjusted linear approximation to solve a large class of dynamic models with time-varying and non-Gaussian risk. Our approach generalizes lognormal affine approximations commonly used in the macro-finance literature and can be seen as a first-order perturbation around the...

We propose a simple risk-adjusted linear approximation to solve a large class of dynamic models with time-varying and non-Gaussian risk. Our approach generalizes lognormal affine approximations commonly used in the macro-finance literature and can be seen as a first-order perturbation around the...

A divide and conquer algorithm for exploiting policy function monotonicity is proposed and analyzed. To solve a discrete problem with n states and n choices, the algorithm requires at most nlog2(n)+5n objective function evaluations. In contrast, existing methods for nonconcave problems require...

Linear Methods are often used to compute approximate solutions to dynamic models, as these models often cannot be solved analytically. Linear methods are very popular, as they can easily be implemented. Also, they provide a useful starting point for understanding more elaborate numerical...