|
on Econometrics |
| By: | Shakeeb Khan; Denis Nekipelov |
| Abstract: | This paper explores the uniformity of inference for parameters of interest in nonlinear models with endogeneity. The notion of uniformity is fundamental in these models because due to potential endogeneity, the behavior of standard estimators of these parameters is shown to vary with where they lie in the parameter space. Consequently, uniform inference becomes nonstandard in a fashion that is loosely analogous to inference complications found in the unit root and weak instruments literature, as well as the models recently studied in Andrews and Cheng (2012a), Andrews and Cheng (2012b) and Chen, Ponomareva, and Tamer (2011). We illustrate this point with two models widely used in empirical work. The first is the standard sample selection model, where the parameter is the intercept term (Heckman (1990), Andrews and Schafgans (1998) and Lewbel (1997a)). We show that with selection on unobservables, asymptotic theory for this parameter is not standard in terms of there being nonparametric rates and non-gaussian limiting distributions. In contrast if the selection is on observables only, rates and asymptotic distribution are standard, and consequently an inference method that is uniform to both selection on observables and unobservables is required. As a second example, we consider the well studied treatment effect model in program evaluation (Rosenbaum and Rubin (1983) and Hirano, Imbens, and Ridder (2003)), where a parameter of interest is the ATE. Asymptotic behavior for existing estimators varies between standard and nonstandard across differing levels of treatment heterogeneity, thus also requiring new inference methods. |
| Keywords: | Selection on observables and unobservables, uniform inference, fixed and drifting sequences of parameters |
| JEL: | C12 C13 C14 C15 |
| Date: | 2013 |
| URL: | http://d.repec.org/n?u=RePEc:duk:dukeec:13-16&r=ecm |
| By: | Bai, Jushan |
| Abstract: | This paper considers dynamic panel models with a factor error structure that is correlated with the regressors. Both short panels (small T) and long panels (large T) are considered. With a small T, consistent estimation requires either a suitable formulation of the reduced form or an appropriate conditional equation for the first observation. Also needed is a suitable control for the correlation between the effects and the regressors. Under the factor error structure, the panel system implies parameter constraints between the mean vector and the covariance matrix. We explore the constraints through a quasi-FIML approach. The factor process is treated as parameters and it can have arbitrary dynamics under both fixed and large T. The large T setting involves incidental parameters because the number of parameters (including the time effects, the factor process, the heteroskedasticity parameters) increases with T. Even though an increasing number of parameters are estimated, we show that there is no incidental parameters bias to affect the limiting distributions; the estimator is centered at zero even scaled by the fast convergence rate of root-NT. We also show that the quasi-FIML approach is efficient under both fixed and large T, despite non-normality, heteroskedasticity, and incidental parameters. Finally we develop a feasible and fast algorithm for computing the quasi-FIML estimators under interactive effects. |
| Keywords: | factor structure, interactive effects, incidental parameters, predetermined regressors, heterogeneity and endogeneity, quasi-FIML, efficiency |
| JEL: | C1 C3 C31 |
| Date: | 2013–09 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:50267&r=ecm |
| By: | Thomas Lux |
| Abstract: | This paper shows how exact solutions for the transient density of a large class of continuous-time Markov switching models can be obtained. We illustrate the pertinent approach for both simple diffusion models with a small number of regimes as well as for the more complicated so-called Poisson multifractal model introduced by Calvet and Fisher (2001) with an arbitrarily large number of regimes. Our results can be immediately applied as well to various popular Markov switching models in financial economics. Closed-form solutions provide for the possibility of exact maximum likelihood estimation for discretely sampled Markov-switching diffusions and also facilitate the use of such models in applied tasks such as option pricing and portfolio management |
| Keywords: | regime switching, continuous-time models, multifractal models |
| JEL: | C13 C58 G12 |
| Date: | 2013–09 |
| URL: | http://d.repec.org/n?u=RePEc:kie:kieliw:1871&r=ecm |
| By: | Maria Kazakova (Russian Presidential Academy of National Economy and Public Administration) |
| Abstract: | In estimating production functions of the researcher is faced with a number of problems, which in the most general form, can be divided into two groups: the problems associated with most econometric estimation techniques, and problems of measurement factors. In the econometric estimation of production functions of the two main problems are linearization of the functional forms and elimination of endogenous explanatory variables. The latter problem is specific to mikroproizvodstvennyh functions, so it we discuss in more detail. Using statistics available in most cases leads to an inaccurate measurement of the volume and cost of factors of production firm that can shift the assessments of the technical characteristics of the production functions. In this paper we detail the mechanism and causes of such bias Example mikroproizvodstvennyh functions. Particular attention is paid to such specific microeconomic production functions as the problem of endogeneity of explanatory variables. Microfunctions also applied to the problems connected with measuring the volume of output and factor costs. |
| Keywords: | production functions |
| Date: | 2013–05 |
| URL: | http://d.repec.org/n?u=RePEc:rnp:wpaper:32&r=ecm |
| By: | Sessi Tokpavi |
| Abstract: | We introduce in this paper a testing approach that allows checking whether two financial institutions are systemically equivalent, with systemic risk measured by CoVaR (Adrian and Brunnermeier, 2011). The test compares the difference in CoVaR forecasts for two financial institutions via a suitable loss function that has an economic content. Our testing approach differs from those in the literature in the sense that it is conditional, and helps evaluating in a forward-looking manner, the extent to which statistically significant differences in CoVaR forecasts can be attributed to lag values of market state variables. Moreover, the test can be used to identify systemically important financial institutions (SIFIs). Extensive Monte Carlo simulations show that the test has desirable small sample properties. With an application on a sample including 70 large U.S. financial institutions, our conditional test using market state variables such as VIX and various yield spreads, reveals more (resp. less) heterogeneity in the systemic profiles of these institutions compared to its unconditional version, in crisis (resp. non-crisis) period. It also emerges that the systemic ranking provided by our testing approach is a good forecast of a financial institution's sensitivity to a crisis. This is in contrast to the ranking obtained directly using CoVaR forecasts which has less predictive power because of estimation uncertainty. |
| Keywords: | Systemic Risk, SIFIs, CoVaR, Estimation Uncertainty, Conditional Predictive Ability Test. |
| JEL: | G32 C53 C58 |
| Date: | 2013 |
| URL: | http://d.repec.org/n?u=RePEc:drm:wpaper:2013-27&r=ecm |
| By: | Michael J. Dickstein; Eduardo Morales |
| Abstract: | Many economic decisions involve a binary choice - for example, when consumers decide to purchase a good or when firms decide to enter a new market. In such settings, agents’ choices often depend on imperfect expectations of the future payoffs from their decision (expectational error) as well as factors that the econometrician does not observe (structural error). In this paper, we show that expectational error, under an assumption of rational expectations, is a source of classical measurement error, and we propose a novel moment inequality estimator that accounts for both expectational error and structural error in a binary choice model. With simulated data and Chilean firm-level customs data, we illustrate the identifying power of our inequalities and show the biases that arise when one ignores either source of error. We use the customs data to estimate the fixed costs exporters face when entering a new market. |
| JEL: | C14 C25 F14 L10 |
| Date: | 2013–09 |
| URL: | http://d.repec.org/n?u=RePEc:nbr:nberwo:19486&r=ecm |
| By: | Keith Head; Thierry Mayer |
| Abstract: | This chapter focuses on the estimation and interpretation of gravity equations for bilateral trade. This necessarily involves a careful consideration of the theoretical underpinnings since it has become clear that naive approaches to estimation lead to biased and frequently misinterpreted results. There are now several theory-consistent estimation methods and we argue against sole reliance on any one method and instead advocate a toolkit approach. One estimator may be preferred for certain types of data or research questions but more often the methods should be used in concert to establish robustness. In recent years, estimation has become just a first step before a deeper analysis of the implications of the results, notably in terms of welfare. We try to facilitate diffusion of best-practice methods by illustrating their application in a step-by-step cookbook mode of exposition. |
| Keywords: | Gravity equations;International trade |
| JEL: | F10 |
| Date: | 2013–09 |
| URL: | http://d.repec.org/n?u=RePEc:cii:cepidt:2013-27&r=ecm |