Another Look to the Price-Dividend Ratio: A Markov-Switching Approach

This paper analyzes the stationarity of this ratio in the context of a Markov-switching model à la Hamilton (1989) where an asymmetric speed of adjustment is introduced. This particular specification robustly supports a nonlinear reversion process and identifies two relevant episodes: the post-war period from the mid-50’s to the mid-70’s and the so called “90’s boom” period. A three-regime Markov-switching model displays the best regime identification and reveals that only the first part of the 90’s boom (1985-1995) and the post-war period are near-nonstationary states. Interestingly, the last part of the 90’s boom (1996-2000), characterized by a growing price-dividend ratio, is entirely attributed to a regime featuring a highly reverting process.

Unemployment rates and population changes in Spain

Published as an article in: Journal of Applied Economics, 2004, vol. VII, pages 47-76.

An Alternative View of the US Price-Dividend Ratio Dynamics

As a necessary condition for the validity of the present value model, the price-dividend ratio must be stationary. However, significant market episodes seem to provide evidence of prices significantly drifting apart from dividends while other episodes show prices anchoring back to dividends. This paper investigates the stationarity of this ratio in the context of a Markov- switching model à la Hamilton (1989) where an asymmetric speed of adjustment towards a unique attractor is introduced. A three-regime model displays the best regime identification and reveals that the first part of the 90’s boom (1985-1995) and the post-war period are characterized by a stationary state featuring a slow reverting process to a relatively high attractor. Interestingly, the latter part of the 90’s boom (1996-2000), characterized by a growing price-dividend ratio, is entirely attributed to a stationary regime featuring a highly reverting process to the attractor. Finally, the post-Lehman Brothers episode of the subprime crisis can be classified into a temporary nonstationary regime.

Learning Multirobot Hose Transportation and Deployment by Distributed Round-Robin Q-Learning

Multi-Agent Reinforcement Learning (MARL) algorithms face two main difficulties: the curse of dimensionality, and environment non-stationarity due to the independent learning processes carried out by the agents concurrently. In this paper we formalize and prove the convergence of a Distributed Round Robin Q-learning (D-RR-QL) algorithm for cooperative systems. The computational complexity of this algorithm increases linearly with the number of agents. Moreover, it eliminates environment non sta tionarity by carrying a round-robin scheduling of the action selection and execution. That this learning scheme allows the implementation of Modular State-Action Vetoes (MSAV) in cooperative multi-agent systems, which speeds up learning convergence in over-constrained systems by vetoing state-action pairs which lead to undesired termination states (UTS) in the relevant state-action subspace. Each agent's local state-action value function learning is an independent process, including the MSAV policies. Coordination of locally optimal policies to obtain the global optimal joint policy is achieved by a greedy selection procedure using message passing. We show that D-RR-QL improves over state-of-the-art approaches, such as Distributed Q-Learning, Team Q-Learning and Coordinated Reinforcement Learning in a paradigmatic Linked Multi-Component Robotic System (L-MCRS) control problem: the hose transportation task. L-MCRS are over-constrained systems with many UTS induced by the interaction of the passive linking element and the active mobile robots.