A new perspective on the exporter productivity premium: online trade

We use a unique firm-level data set including 9000 companies from 26 European Union countries covering four different sectors to take a close look at the relationship between online exports and productivity. The online exporter productivity premium is estimated using different techniques (ordinary least squares, quantile regressions and robust estimation). Results consistently indicate that the estimated online exporter productivity premium is statistically different from zero, positive and significant from an economic point of view. European online exporters, according to these results, are approximately 2% more productive than non-online exporters. Productivity differences between firms could be related to variables that are not included in the empirical model. More research would be needed to address this issue in the future.     

PITFALLS IN CROSS‐SECTION STUDIES WITH INTEGRATED REGRESSORS: A SURVEY AND NEW DEVELOPMENTS

In cross‐section studies, if the dependent variable is I(0) but the regressor is I(1), the true slope must be zero in the resulting “unbalanced regression.” A spuriously significant relationship may be found in large cross‐sections, however, if the integrated regressor is related to a stationary variable that enters the DGP but is omitted from the regression. The solution is to search for the related stationary variable, in some cases the first difference of the integrated regressor, in other cases, a categorical variable that can take on limited number of values which depend on the integrated variable. We present an extensive survey, new developments, and applications particularly in finance.     

Predictive regressions under asymmetric loss: Factor augmentation and model selection

This paper discusses the specifics of forecasting using factor-augmented predictive regressions under general loss functions. In line with the literature, we employ principal component analysis to extract factors from the set of predictors. In addition, we also extract information on the volatility of the series to be predicted, since the volatility is forecast-relevant under non-quadratic loss functions. We ensure asymptotic unbiasedness of the forecasts under the relevant loss by estimating the predictive regression through the minimization of the in-sample average loss. Finally, we select the most promising predictors for the series to be forecast by employing an information criterion that is tailored to the relevant loss. Using a large monthly data set for the US economy, we assess the proposed adjustments in a pseudo out-of-sample forecasting exercise for various variables. As expected, the use of estimation under the relevant loss is found to be effective. Using an additional volatility proxy as the predictor and conducting model selection that is tailored to the relevant loss function enhances the forecast performance significantly.     

The Local Whittle Estimator of Long-Memory Stochastic Volatility

We propose a new semiparametric estimator of the degree of persistencein volatility for long memory stochastic volatility (LMSV) models.The estimator uses the periodogram of the log squared returnsin a local Whittle criterion which explicitly accounts for thenoise term in the LMSV model. Finite-sample and asymptotic standarderrors for the estimator are provided. An extensive simulationstudy reveals that the local Whittle estimator is much lessbiased and that the finite-sample standard errors yield moreaccurate confidence intervals than the widely-used GPH estimator.The estimator is also found to be robust against possible leverageeffects. In an empirical analysis of the daily Deutsche Mark/USDollar exchange rate, the new estimator indicates stronger persistencein volatility than the GPH estimator, provided that a largenumber of frequencies is used.     

Time Inhomogeneous Multiple Volatility Modeling

Price variations at speculative markets exhibit positive autocorrelationand cross correlation. Due to large parameter spaces necessaryfor joint modeling of variances and covariances, multivariateparametric volatility models become easily intractable in practice.We propose an adaptive procedure that identifies periods ofsecond-order homogeneity for each moment in time. To overcomethe high dimensionality of the problem we transform the multivariateseries into a set of univariate processes. We discuss thoroughlythe implementation of the adaptive technique. Theoretical andMonte Carlo results are given. We provide two applications ofthe new method. For a bivariate exchange rate series we comparethe multivariate GARCH approach with our method and find thelatter to be more in line with the underlying assumption ofindependently distributed innovations. Analyzing a 23-dimensionalvector of asset returns we underscore the case for adaptivemodeling in high-dimensional systems.     

Beyond Single-Factor Affine Term Structure Models

This article proposes a new approach to testing for the hypothesisof a single priced risk factor driving the term structure ofinterest rates. The method does not rely on any parametric specificationof the state variable dynamics or the market price of risk.It simply exploits the constraint imposed by the no-arbitragecondition on instantaneous expected bond returns. In order toachieve our goal, we develop a Kolmogorov-Smirnov test and applyit to data on Treasury bills and bonds for both the United Statesand Spain. We find that the single risk factor hypothesis cannotbe rejected for either dataset.     

Economic design of one-sided screening procedures based on a correlated variable with all parameters unknown

Summary Economic design of one-sided screening procedures for improving outgoing product quality based on a screening variable is considered for the case with all parameters unknown. It is assumed that the performance and screening variables are jointly normally distributed and that costs are incurred by screening inspection, acceptance of an imperfect item, and disposition of a rejected one. A method for finding optimal cutoff value based on the predictive distribution is presented.     

Option Trading and the Intervalling Effect Bias in Beta

Prior studies show that the beta coefficient of a security changes systematically as the length of measurement interval is varied. This phenomenon, which is called the intervalling effect bias in beta, has been attributed to the friction in the trading system that causes the delays in the price-adjustment process. This study shows that option listing is associated with a decline in the beta intervalling effect bias. The decline is most pronounced for small firms. We also find that our sample firms grow significantly after option listing. Since prior research indicates that market value is a major determinant of the magnitude of the intervalling effect, we re-examine our results using a subsample that controls for market value. The results indicate that the decline in the beta bias from the pre-listing to post-listing period is still prevalent after we control for the change in firm size. Overall, the evidence is consistent with the notion that option trading reduces the delays in the price-adjustment process, which in turn reduces the intervalling effect bias in beta.     

LABOUR SUPPLY ESTIMATION AND PUBLIC POLICY

Abstract. Economists devote considerable energies towards refining their econometric techniques to overcome difficulties connected with conducting empirical research. Despite advances in technique. it is not clear whether further refinement in this direction is worthwhile for policy purposes. It may be that no further amount of statistical adjustment of inadequate data will increase understanding, and that better data is simply necessary to add to our knowledge. But rarely is sufficient credit paid to new forms of data. In short, econometric technique is emphasized to the neglect of data innovation, as if new data were merely lying about waiting for an ingenious suggestion for use. This paper surveys advances of the last twenty five years in estimating labour supply for policy purposes with a view towards appreciating the relative contribution of both improvements in econometric technique as well as developments of new data.
After briefly detailing the key parameters which economists have sought to estimate, we describe the early 'first generation' research (circa 1970), which is plagued by problems of unobservable variables, measurement errors, truncation and selectivity bias, and non linear budget constraints. 'Second generation' research constitute attempts to resolve one or more of these difficulties, and the respective contribution of econometric technique and new data is acknowledged and assessed, including the contribution of data generated by large scale social experiments in which participants are randomly assigned to different guaranteed income plans and their labour supply behaviour measured.