
on Econometrics 
By:  Hiroaki Chigira; Taku Yamamoto 
Abstract:  This paper is concerned with the estimation of the autoregressive parameter of dynamic panel data models. We propose a biascorrected GMM estimator whose bias is smaller than that of many existing GMM estimators. And we propose a small sample corrected estimator of the variance in order to reduce the size distortion of the Wald test. These estimators are easy to calculate and do not require preliminary estimates. The Monte Carlo experiments indicate that in terms of both bias and size distortion, the bias corrected estimator out performs Blundell and Bond's (1998) system estimator even when using Windmeijer's (2005) correction of the estimated variance of the system estimator. 
Keywords:  Generalized method of moments, bias correction, panel data 
JEL:  C12 C23 
Date:  2006–07 
URL:  http://d.repec.org/n?u=RePEc:hst:hstdps:d06177&r=ecm 
By:  Elena Pesavento 
Abstract:  Numerous tests for integration and cointegration have been proposed in the literature. Since Elliott, Rothemberg and Stock (1996) the search for tests with better power has moved in the direction of finding tests with some optimality properties both in univariate and multivariate models. Although the optimal tests constructed so far have asymptotic power that is indistinguishable from the power envelope, it is well known that they can have severe size distortions in finite samples. This paper proposes a simple and powerful test that can be used to test for unit root or for no cointegration when the cointegration vector is known. Although this test is not optimal in the sense of Elliott and Jansson (2003), it has better finite sample size properties while having asymptotic power curves that are indistinguishable from the power curves of optimal tests. Similarly to Hansen (1995), Elliott and Jansson (2003), Zivot (2000), and Elliott, Jansson and Pesavento (2005) the proposed test achieves higher power by using additional information contained in covariates correlated with the variable being tested. The test is constructed by applying Hansen's test to variables that are detrended under the alternative in a regression augmented with leads and lags of the stationary covariates. Using local to unity parametrization, the asymptotic distribution of the test under the null and the local alternative is analytically computed. 
Date:  2006–07 
URL:  http://d.repec.org/n?u=RePEc:emo:wp2003:0606&r=ecm 
By:  Ralph D. Snyder; Anne B. Koehler 
Abstract:  It is a common practice to complement a forecasting method such as simple exponential smoothing with a monitoring scheme to detect those situations where forecasts have failed to adapt to structural change. It will be suggested in this paper that the equations for simple exponential smoothing can be augmented by a common monitoring statistic to provide a method that automatically adapts to structural change without human intervention. It is shown that the resulting equations conform to those of damped trend corrected exponential smoothing. In a similar manner, exponential smoothing with drift, when augmented by the same monitoring statistic, produces equations that split the trend into long term and short term components. 
Keywords:  Forecasting, exponential smoothing, tracking signals. 
JEL:  C32 
Date:  2006–08 
URL:  http://d.repec.org/n?u=RePEc:msh:ebswps:200616&r=ecm 
By:  L. Israelsen; J. McDonald 
Abstract:  A model for measurement error is developed, based on the assumption that measurement error is random, multiplicative, and independent of the level of actual income. Thus, measured income is defined as the product of actual income and measurement error. Flexible parametric forms are utilized to model the distributions of actual income (generalized gamma) and measurement error (inverse generalized gamma). The probability density of measured income is then derived as a generalized beta of the second kind (GB2). Estimation of the parameters of the GB2 (measured income), then allows an estimate to be made of the pdf of actual income, from which corresponding estimated means, variances, Gini coefficients, and Lorenz curves are obtained. An identification problem in the parameters of actual and measured income is solved with additional information as to the average fraction of actual income reported. The implied characteristics of measurement error are also obtained. The procedure is applied to income data from several Latin American economies, and estimates of actual income distribution characteristics are derived from measured income. It is found, in some cases, that when measured income inequality moved in one direction over time, actual income inequality moved in the opposite direction. This finding has important implications for the evaluation of policies designed to affect relative equality in the distribution of income. 
URL:  http://d.repec.org/n?u=RePEc:usu:wpaper:200018&r=ecm 
By:  In Choi; Timothy K. Chue 
Abstract:  We develop subsamplingbased tests of stockreturn predictability and apply them to U.S. data. These tests allow for multiple predictor variables with localtounit roots. By contrast, previous methods that model the predictor variables as nearly integrated are only applicable to univariate predictive regressions. Simulation results demonstrate that our subsamplingbased tests have desirable size and power properties. Using stockmarket valuation ratios and the riskfree rate as predictors, our univariate tests show that the evidence of predictability is more concentrated in the 19261994 subperiod. In bivariate tests, we find support for predictability in the full sample period 19262004 and the 19522004 subperiod as well. For the subperiod 19522004, we also consider a number of consumptionbased variables as predictors for stock returns and find that they tend to perform better than the dividendprice ratio. Among the variables we consider, the predictive power of the consumptionwealth ratio proposed by Lettau and Ludvigson (2001a, 2001b) seems to be the most robust. Among variables based on habit persistence, Campbell and Cochrane's (1999) nonlinear specication tends to outperform a more traditional, linear specification. 
Keywords:  Subsampling, localtounit roots, predictive regression, stockreturn predictability, consumptionbased models 
Date:  2006–07 
URL:  http://d.repec.org/n?u=RePEc:hst:hstdps:d06178&r=ecm 
By:  Niels Framroze Møller (Department of Economics, University of Copenhagen) 
Abstract:  This paper attempts to clarify the connection between simple economic theory models and the approach of the Cointegrated VectorAutoRegressive model (CVAR). By considering (stylized) examples of simple static equilibrium models, it is illustrated in detail, how the theoretical model and its structure and assumptions can be translated into a CVAR. We also see how the CVAR allows for explicit hypotheses about transitory dynamics, that could be relevant for assessing price rigidity, and hence, "the length of the short run"  a controversial issue in traditional macroeconomics. Moreover, it is demonstrated how other controversial hypotheses such as Rational Expectations can be formulated directly as restrictions on the CVARparameters. A simple example of a "Neoclassical synthetic" ASAD model is also formulated. Finally, the partial general equilibrium distinction is related to the CVAR as well. Further fundamental extensions and advances to more sophisticated theory models, such as those related to dynamics and expectations (in the structural relations) are left for future papers. 
Keywords:  cointegrated VAR; static theory models; ASAD; price rigidities; rational expectations; general equilibrium 
JEL:  C32 
Date:  2006–06 
URL:  http://d.repec.org/n?u=RePEc:kud:kuiedp:0615&r=ecm 
By:  Aldrich, John 
Abstract:  This paper considers J. M. Keynes as a statistician and philosopher of statistics and the reaction of English statisticians to his critique of their work. It follows the development of Keynes's thinking through the two versions of his fellowship dissertation The Principles of Probability (1907/8) to his book A Treatise on Probability (1921). It places Keynes's ideas in the context of contemporary English and Continental statistical thought. Of the statisticians considered special attention is paid to the reactions of .four: Edgeworth, Bowley, Jeffreys and R. A. Fisher. Keywords; Keynes, Edgeworth, Bowley, Pearson, Jeffreys, Fisher, Lexis, Bortkiewicz. JEL Classification: B16 B23 B30 
URL:  http://d.repec.org/n?u=RePEc:stn:sotoec:0611&r=ecm 