The term structure of interest rates or yield curve is a function relating the interest rate with its own term. Nonlinear regression models of Nelson - Si egel and Svensson were used to estimate the yield curve using a sample of historical data supplied by th e National Stock Exchange of Costa Rica. The optimization problem involved in the estimation process of model parameters is addressed by the use of four well known combinatorial optimization metaheu-ristics: Ant colony optimization, Genetic algorithm, Part icle swarm optimization and Simulated annealing. The aim of the study is to improve the local minima obtained by a classical quasi - Newton optimization m ethod using a descent direction. Good results with at least two metaheuristics are achieved, Particle sw arm optimization and Simulated annealing.