
on Econometrics 
By:  Chen, Willa; Deo, Rohit 
Abstract:  The restricted likelihood, which has small curvature, is derived for the bivariate predictive regression problem. The bias of the Restricted Maximum Likelihood (REML) estimates is shown to be approximately 50% less than that of the OLS estimates near the unit root, without loss of efficiency. The error in the chisquare approximation to the distribution of the REML based Likelihood Ratio Test (RLRT) for no predictability is shown to be (3/4ρ²)(1/n)(G₃(.)G₁(.))+O(1/n^2), where ρ<1 is the correlation of the innovation series and G_{s}(.) is the c.d.f. of a χ_{s}² random variable. This very small error, free of the AR parameter, implies that the RLRT for predictability has very good size properties even when the regressor is nearly integrated, a fact borne out by our simulation study. The Bartlett corrected RLRT achieves an O(1/n^2) error. The power of the RLRT under a sequence of local Pitman alternatives is obtained and shown to be always identical to that of the Wald test and higher than that of the Rao score test for empirically relevant regions. Some extensions to the case of vector AR(1) regressors and more general univariate regressors are provided. The RLRT is found to work very well in simulations and to be robust to nonnormal errors. 
Keywords:  Bartlett correction; likelihood ratio test; curvature 
JEL:  C13 C12 
Date:  2006 
URL:  http://d.repec.org/n?u=RePEc:pra:mprapa:82&r=ecm 
By:  Matheson, Troy D 
Abstract:  This paper focuses on forecasting four key New Zealand macroeconomic variables using a dynamic factor model and a large number of predictors. We compare the (simulated) realtime forecasting performance of the factor model with a variety of other timeseries models (including the Reserve Bank of New Zealand’s published forecasts), and we gauge the sensitivity of our results to alternative variableselection algorithms. We find that the factor model performs particularly well at longer horizons. 
JEL:  G00 G0 
Date:  2006–04–13 
URL:  http://d.repec.org/n?u=RePEc:pra:mprapa:807&r=ecm 
By:  Souza, Leonardo 
Abstract:  This paper aims at showing that a temporal aggregation and a specific bandwidth reduction lead to the same asymptotic properties in estimating long memory by Geweke and PorterHudak’s (1983) and Robinson’s (1995b) estimators (henceforth GPH and GSPR). In other words, irrespectively of the level of temporal aggregation, the asymptotic properties of the estimator are uniquely determined by the number of periodogram ordinates used in the estimation, provided some mild additional assumptions are imposed. Monte Carlo simulations show that this result is a good approximation in finite samples. A real example with the daily US Dollar/French Franc exchange rate series is also provided. 
Keywords:  Temporal Aggregation; Long Memory; Bandwidth; Spectrum 
JEL:  C10 
Date:  2006–09 
URL:  http://d.repec.org/n?u=RePEc:pra:mprapa:108&r=ecm 
By:  Frimpong, Joseph Magnus; OtengAbayie, Eric Fosu 
Abstract:  This paper models and forecasts volatility (conditional variance) on the Ghana Stock Exchange using a random walk (RW), GARCH(1,1), EGARCH(1,1), and TGARCH(1,1) models. The unique ‘three days a week’ Databank Stock Index (DSI) is used to study the dynamics of the Ghana stock market volatility over a 10year period. The competing volatility models were estimated and their specification and forecast performance compared with each other, using AIC and LL information criteria and BDS nonlinearity diagnostic checks. The DSI exhibits the stylized characteristics such as volatility clustering, leptokurtosis and asymmetry effects associated with stock market returns on more advanced stock markets. The random walk hypothesis is rejected for the DSI. Overall, the GARCH (1,1) model outperformed the other models under the assumption that the innovations follow a normal distribution. 
Keywords:  Ghana Stock Exchange; developing financial markets; volatility; GARCH model 
JEL:  C52 G15 G10 C22 
Date:  2006–10–07 
URL:  http://d.repec.org/n?u=RePEc:pra:mprapa:593&r=ecm 
By:  Lu, Jingfeng; Perrigne, Isabelle 
Abstract:  Estimating bidders’ risk aversion in auctions is a challeging problem because of identification issues. This paper takes advantage of bidding data from two auction designs to identify nonparametrically the bidders’ utility function within a private value framework. In particular, ascending auction data allow us to recover the latent distribution of private values, while firstprice sealedbid auction data allow us to recover the bidders’ utility function. This leads to a nonparametric estimator. An application to the US Forest Service timber auctions is proposed. Estimated utility functions display concavity, which can be partly captured by constant relative risk aversion. 
Keywords:  Risk Aversion; Nonparametric Identi.cation; Nonparametric and Semiparametric Estimation; Timber Auctions. 
JEL:  D44 C14 
Date:  2006–11 
URL:  http://d.repec.org/n?u=RePEc:pra:mprapa:948&r=ecm 
By:  Ghent, Andra 
Abstract:  I generate priors for a VAR from four competing models of economic fluctuations: a standard RBC model, Fisher’s (2006) investmentspecific technology shocks model, an RBC model with capital adjustment costs and habit formation, and a sticky price model with an unaccommodating monetary authority. I compare the accuracy of the forecasts made with each of the resulting VARs. The economic models generate similar forecast errors to one another. However, at horizons of one to two years and greater, the models generally yield superior forecasts to those made using both an unrestricted VAR and a VAR that uses shrinkage from a Minnesota prior. 
Keywords:  Model Evaluation; Priors from DSGE models; Economic Fluctuations; Hours Debate; Business Cycles; 
JEL:  C52 E37 E32 C53 E3 C11 
Date:  2006–08 
URL:  http://d.repec.org/n?u=RePEc:pra:mprapa:180&r=ecm 
By:  Mishra, SK 
Abstract:  In this paper Sato’s twolevel CES production function has been estimated by nonlinear regression carried out through five different methods of optimization, namely, the HookeJeeves Pattern Moves (HJPM), the HookeJeevesQuasiNewton (HJQN), the RosenbrockQuasiNewton (RQN), the Differential Evolution (DE) and the Repulsive Particle Swarm methods (RPS). The last two methods are particularly suited to optimization of extremely nonlinear (often multimodal) objective functions. While data may be containing outliers, the method of least squares has a clear disadvantage as it may be pulled by extremely small or large errors. The absolute deviation estimation of parameters is more suitable in such cases. This paper has made an attempt to estimation of parameters of Sato’s twolevel CES production function by minimizing the sum of absolute errors. The minimization has been done by the five methods noted above. While the HJPM and the HJQN perform poorly at minimizing the sum of absolute deviations, the RQN performs much better. The DE and the RPS perform very well in estimating the parameters. 
Keywords:  Sato’s productions function; CES; constant elasticity of substitution; twolevel; nonlinear regression; Hooke Jeeves; Pattern move; QuasiNewton; Rosenbrock; Repulsive Particle swarm; Differential Evolution; Global Optimization; Econometrics; Estimation; Outliers; Least absolute deviation; error 
JEL:  C20 D20 C61 
Date:  2006–11–26 
URL:  http://d.repec.org/n?u=RePEc:pra:mprapa:932&r=ecm 