Abstract: Although economic processes and systems are in general simple in nature, the underlying dynamics are complicated and seldom understood. Recognizing this, in this paper we use a nonstationary-conditional Markov process model of observed aggregate data to learn about and recover causal influence information associated with the underlying dynamic micro-behavior. Estimating equations are used as a link to the data and to model the dynamic conditional Markov process. To recover the unknown transition probabilities, we use an information theoretic approach to model the data and derive a new class of conditional Markov models. A quadratic loss function is used as a basis for selecting the optimal member from the family of possible likelihood-entropy functional(s). The asymptotic properties of the resulting estimators are demonstrated, and a range of potential applications is discussed.
Abstract: The classical Chow test for structural instability requires strictly exogenousregressors and a break-point specified in advance. In this paper, we consider twogeneralisations, the one-step recursive Chow test (based on the sequence of studentisedrecursive residuals) and its supremum counterpart, which relaxes these requirements. We useresults on the strong consistency of regression estimators to show that the one-step test isappropriate for stationary, unit root or explosive processes modelled in the autoregressivedistributed lags (ADL) framework. We then use the results in extreme value theory to developa new supremum version of the test, suitable for formal testing of structural instability withan unknown break-point. The test assumes the normality of errors and is intended to be usedin situations where this can be either assumed nor established empirically. Simulations showthat the supremum test has desirable power properties, in particular against level shifts latein the sample and against outliers. An application to U.K. GDP data is given.
Abstract: The vast majority of spatial econometric research relies on the assumption that the spatial network structure is known a priori. This study considers a two-step estimation strategy for estimating the n(n-1) interaction effects in a spatial autoregressive panel model where the spatial dimension is potentially large. The identifying assumption is approximate sparsity of the spatial weights matrix. The proposed estimation methodology exploits the Lasso estimator and mimics two-stage least squares (2SLS) to account for endogeneity of the spatial lag. The developed two-step estimator is of more general interest. It may be used in applications where the number of endogenous regressors and the number of instrumental variables is larger than the number of observations. We derive convergence rates for the two-step Lasso estimator. Our Monte Carlo simulation results show that the two-step estimator is consistent and successfully recovers the spatial network structure for reasonable sample size, T.
Abstract: In this study, I investigate the necessary condition for the consistency of the maximum likelihood estimator (MLE) of spatial models with a spatial moving average process in the disturbance term. I show that the MLE of spatial autoregressive and spatial moving average parameters is generally inconsistent when heteroskedasticity is not considered in the estimation. I also show that the MLE of parameters of exogenous variables is inconsistent and determine its asymptotic bias. I provide simulation results to evaluate the performance of the MLE. The simulation results indicate that the MLE imposes a substantial amount of bias on both autoregressive and moving average parameters.
Abstract: As a basis for information recovery in open dynamic microeconomic systems, we emphasize the connection between adaptive intelligent behavior, causal entropy maximization and self-organized equilibrium seeking behavior. This entropy-based causal adaptive behavior framework permits the use of information-theoretic methods as a solution basis for the resulting pure and stochastic inverse economic-econometric problems. We cast the information recovery problem in the form of a binary network and suggest information-theoretic methods to recover estimates of the unknown binary behavioral parameters without explicitly sampling the configuration-arrangement of the sample space.
Abstract: This paper focuses on finding starting-values for the estimation of Vector STAR models. Based on a Monte Carlo study, different procedures are evaluated. Their performance is assessed with respect to model fit and computational effort. I employ (i) grid search algorithms and (ii) heuristic optimization procedures, namely differential evolution, threshold accepting, and simulated annealing. In the equation-by-equation starting-value search approach the procedures achieve equally good results. Unless the errors are cross-correlated, equation-by-equation search followed by a derivative-based algorithm can handle such an optimization problem sufficiently well. This result holds also for higher-dimensional Vector STAR models with a slight edge for heuristic methods. For more complex Vector STAR models which require a multivariate search approach, simulated annealing and differential evolution outperform threshold accepting and the grid search.