Search results

This article surveys recent developments in the evaluation of point and density forecasts in the context of forecasts made by vector autoregressions. Specific emphasis is placed on highlighting those parts of the existing literature that are applicable to direct multistep forecasts and those parts that are applicable to iterated multistep forecasts. This literature includes advancements in the evaluation of forecasts in population (based on true, unknown model coefficients) and the evaluation of forecasts in the finite sample (based on estimated model coefficients). The article then examines in Monte Carlo experiments the finite-sample properties of some tests of equal forecast accuracy, focusing on the comparison of VAR forecasts to AR forecasts. These experiments show the tests to behave as should be expected given the theory. For example, using critical values obtained by bootstrap methods, tests of equal accuracy in population have empirical size about equal to nominal size.

In this paper we use Monte Carlo sampling experiments to examine the properties of pretest estimators in the random parameters logit (RPL) model. The pretests are for the presence of random parameters. We study the Lagrange multiplier (LM), likelihood ratio (LR), and Wald tests, using conditional logit as the restricted model. The LM test is the fastest test to implement among these three test procedures since it only uses restricted, conditional logit, estimates. However, the LM-based pretest estimator has poor risk properties. The ratio of LM-based pretest estimator root mean squared error (RMSE) to the random parameters logit model estimator RMSE diverges from one with increases in the standard deviation of the parameter distribution. The LR and Wald tests exhibit properties of consistent tests, with the power approaching one as the specification error increases, so that the pretest estimator is consistent. We explore the power of these three tests for the random parameters by calculating the empirical percentile values, size, and rejection rates of the test statistics. We find the power of LR and Wald tests decreases with increases in the mean of the coefficient distribution. The LM test has the weakest power for presence of the random coefficient in the RPL model.

Monte Carlo simulations are a very powerful way to demonstrate the basic sampling properties of various statistics in econometrics. The commercial software package Stata makes these methods accessible to a wide audience of students and practitioners. The purpose of this chapter is to present a self-contained primer for conducting Monte Carlo exercises as part of an introductory econometrics course. More experienced econometricians that are new to Stata may find this useful as well. Many examples are given that can be used as templates for various exercises. Examples include linear regression, confidence intervals, the size and power of t-tests, lagged dependent variable models, heteroskedastic and autocorrelated regression models, instrumental variables estimators, binary choice, censored regression, and nonlinear regression models. Stata do-files for all examples are available from the authors' website http://learneconometrics.com/pdf/MCstata/.

Weak separability is an important concept in many fields of economic theory. This chapter uses Monte Carlo experiments to investigate the performance of newly developed nonparametric revealed preference tests for weak separability. A main finding is that the bias of the sequentially implemented test for weak separability proposed by Fleissig and Whitney (2003) is low. The theoretically unbiased Swofford and Whitney test (1994) is found to perform better than all sequentially implemented test procedures but is found to suffer from an empirical bias, most likely because of the complexity in executing the test procedure. As a further source of information, we also perform sensitivity analyses on the nonparametric revealed preference tests. It is found that the Fleissig and Whitney test seems to be sensitive to measurement errors in the data.

Most spatial econometrics work focuses on spatial dependence in the regressand or disturbances. However, Lesage and Pace (2009) as well as Pace and LeSage2009 showed that the bias in β from applying OLS to a regressand generated from a spatial autoregressive process was exacerbated by spatial dependence in the regressor. Also, the marginal likelihood function or restricted maximum likelihood (REML) function includes a determinant term involving the regressors. Therefore, high dependence in the regressor may affect the likelihood through this term. In addition, Bowden and Turkington (1984) showed that regressor temporal autocorrelation had a non-monotonic effect on instrumental variable estimators.

We provide empirical evidence that many common economic variables used as regressors (e.g., income, race, and employment) exhibit high levels of spatial dependence. Based on this observation, we conduct a Monte Carlo study of maximum likelihood (ML), REML and two instrumental variable specifications for spatial autoregressive (SAR) and spatial Durbin models (SDM) in the presence of spatially correlated regressors.

Findings indicate that as spatial dependence in the regressor rises, REML outperforms ML and that performance of the instrumental variable methods suffer. The combination of correlated regressors and the SDM specification provides a challenging environment for instrumental variable techniques.

We also examine estimates of marginal effects and show that these behave better than estimates of the underlying model parameters used to construct marginal effects estimates. Suggestions for improving design of Monte Carlo experiments are provided.

We compare methods to measure comovement in business cycle data using multi-level dynamic factor models. To do so, we employ a Monte Carlo procedure to evaluate model performance for different specifications of factor models across three different estimation procedures. We consider three general factor model specifications used in applied work. The first is a single-factor model, the second a two-level factor model, and the third a three-level factor model. Our estimation procedures are the Bayesian approach of Otrok and Whiteman (1998), the Bayesian state-space approach of Kim and Nelson (1998) and a frequentist principal components approach. The latter serves as a benchmark to measure any potential gains from the more computationally intensive Bayesian procedures. We then apply the three methods to a novel new dataset on house prices in advanced and emerging markets from Cesa-Bianchi, Cespedes, and Rebucci (2015) and interpret the empirical results in light of the Monte Carlo results.

This paper is intended to test the robustness of the fitness of nested GARCH models.

Both Monte‐Carlo simulation data and real‐world data are used in the paper. Likelihood‐family tests are used to test in‐sample fitness, while mean‐squared prediction error is employed for out‐sample prediction tests.

The paper finds that, generally, the parsimonious principle is found to work well for both criteria. However, it is found that conflict exists between the two criteria: in‐sample likelihood‐family tests pay more attention to conditional distributions or are more sensitive to fat tail effects; while the out‐sample criteria focus more on the accuracy of parameter estimation.

The paper shows that complexity does not necessarily mean good fitness; sometimes, the simpler model can fit better, especially for real‐world data.

We formulate generalized maximum entropy estimators for the general linear model and the censored regression model when there is first order spatial autoregression in the dependent variable. Monte Carlo experiments are provided to compare the performance of spatial entropy estimators relative to classical estimators. Finally, the estimators are applied to an illustrative model allocating agricultural disaster payments.

This paper develops a cross-sectionally augmented distributed lag (CS-DL) approach to the estimation of long-run effects in large dynamic heterogeneous panel data models with cross-sectionally dependent errors. The asymptotic distribution of the CS-DL estimator is derived under coefficient heterogeneity in the case where the time dimension (T ) and the cross-section dimension (N ) are both large. The CS-DL approach is compared with more standard panel data estimators that are based on autoregressive distributed lag (ARDL) specifications. It is shown that unlike the ARDL-type estimator, the CS-DL estimator is robust to misspecification of dynamics and error serial correlation. The theoretical results are illustrated with small sample evidence obtained by means of Monte Carlo simulations, which suggest that the performance of the CS-DL approach is often superior to the alternative panel ARDL estimates, particularly when T is not too large and lies in the range of 30–50.