A flexible sequential Monte Carlo algorithm for shape-constrained regression

10/02/2018

∙

by   Kenyon Ng, et al.

∙

0

∙

share

We propose an algorithm that is capable of imposing shape constraints on regression curves, without requiring the constraints to be written as closed-form expressions, nor assuming the functional form of the loss function. Our algorithm, which is based on Sequential Monte Carlo-Simulated Annealing, only relies on an indicator function that assesses whether or not the constraints are fulfilled, thus allowing us to enforce various complex constraints by specifying an appropriate indicator function without altering other parts of the algorithm. We demonstrate our algorithm by fitting rational function models subject to monotonicity and continuity constraints. The algorithm was implemented using R (R Core Team, 2018) and the code is freely available on GitHub.