There and back again – estimating equivalence scales with measurement error

Using income satisfaction data from the German Socio-Economic Panel, we find large differences in the equivalence weight of a partner when it is being estimated by direct and reverse regressions. We argue that neither of the two models will produce consistent estimates when there is stochastic error in satisfaction and measurement error in incomes. We propose a correction of mismeasured incomes using a constructed alternative income measure. The corrected results are relatively close to those obtained from direct regression. We do not find evidence that previous studies, using the direct regression method, severely suffer from measurement error in incomes.     

Economic information versus quality variation in cross-country data

Data quality in the Penn World Tables varies systematically across countries that have different growth rates and are at different stages of economic development, thus introducing measurement error correlated with variables of economic interest. We explore this problem with three examples from the literature, showing that the problem appears to be minor in growth convergence regressions but serious in estimating the effect of income volatility on growth and in a cross-country test of the Permanent Income Hypothesis. The results suggest, at the least, a need for performing appropriate sensitivity tests before drawing conclusions from analyses based on these data. JEL Classification: E21, O47     

Mixed data sampling expectile regression with applications to measuring financial risk

To avoid information loss or measurement error in traditional methods dealing with mixed frequency data, we develop a novel mixed data sampling expectile regression (MIDAS-ER) model to measure financial risk. We construct the MIDAS-ER model by introducing a MIDAS structure into expectile regressions. This enables us to perform an expectile regression on raw mixed frequency data directly. We apply the proposed MIDAS-ER model to estimate two popular financial risk measures, namely, Value at Risk and Expected Shortfall, with both simulated data and four stock indices, and compare the model's performance with those of several popular models. The outstanding performance of our model demonstrates that high-frequency information helps to improve the accuracy of risk measurement. In addition, the numerical results also imply that our model can be a significant tool for risk-averse investors to control risk losses and for financial institutions to implement robust risk management.     

Growth and human capital: good data,good results

We present a new data set for years of schooling across countries for the 1960–2000 period. The series are constructed from the OECD database on educational attainment and from surveys published by UNESCO. Two features that improve the quality of our data with respect to other series, particularly for series in first-differences, are the use of surveys based on uniform classification systems of education over time, and an intensified use of information by age groups. As a result of the improvement in quality, these new series can be used as a direct substitute for Barro and Lee’s (2001; Oxford Economic Papers, 3, 541–563) data in empirical research. In standard cross-country growth regressions we find that our series yield significant coefficients for schooling. In panel data estimates our series are also significant even when the regressions account for the accumulation of physical capital. Moreover, the estimated macro return is consistent with those reported in labour studies. These results differ from the typical findings of the earlier literature and are a consequence of the reduction in measurement error in the series.      

A survey of recent theoretical developments in the econometrics of panel data

This paper surveys some recent developments in panel data analysis. In particular, it focuses on the error component model which is popular in panel data applications, and discusses recent advances in its estimation under heteroscedasticity, serial correlation, and a general variance-covariance matrix. It also surveys the extensions of this model to the seemingly unrelated regressions case, and the simultaneous equations case. The dynamic case and the incomplete panel data case are also considered, as well as a host of other miscellaneous extensions. Prediction with this model is briefly surveyed and alternative tests for this model are reviewed. While the bibliography is not exhaustive, this survey should complement previous surveys and should prove useful for researchers working in this area.     

Seemingly unrelated regressions with spatial error components

This article considers various estimators using panel data seemingly unrelated regressions (SUR) with spatial error correlation. The true data generating process (DGP) is assumed to be SUR with spatial error of the autoregressive or moving average type. Moreover, the remainder term of the spatial process is assumed to follow an error component structure. Both maximum likelihood (ML) and generalized moments (GM) methods of estimation are used. Using Monte Carlo experiments, we check the performance of these estimators and their forecasts under misspecification of the spatial error process, various spatial weight matrices, and heterogeneous versus homogeneous panel data models.     

Moments of cross-sectional stock market returns and the German business cycle

Based on monthly data covering the period from 1987 to 2021, we analyse whether cross-sectional moments of stock market returns may provide information about the future position of the German business cycle. We apply in-sample forecasting regressions with and without leading indicators as control variables, pseudo-out-of-sample exercises, autoregressive distributed lag models, and impulse-response functions estimated by local projections. We find in-sample predictive power of the first and third cross-section moments for the future growth of industrial production, even if one controls for well-established leading indicators for the German business cycle. Out-of-sample tests show that these variables reduce the relative mean squared error compared with benchmark models. We do not find a long-run relation between the moment series and industrial production. The dynamic response of industrial production to a shock on the cross-section moments is in line with the other results.     

EXPORTS,EXPORT COMPOSITION AND GROWTH: A SIMULTANEOUS ERROR-CORRECTION MODEL FOR TUNISIA

The export-economic growth relationship has been dealt with in several empirical studies concerning cross-section and time series data. Although this research has contributed to measure the impact of exports on economic growth, it still has three main drawbacks. First, the studies on cross-section data suppose homogeneous production techniques for the countries involved. Second, most of the research introduces a bias in the estimation of the export impact on economic growth since it neglects all effects of simultaneity between these two variables. Finally, most of the estimated regressions in these studies may constitute ‘spurious‘ regressions since the analysis of the stationarity of the variables is missing. This paper considers the Tunisian case and tries to study the dynamics between growth and exports through a simultaneous error correction model. The study shows the presence of a positive and significant relationship between exports and economic growth driven by manufactured exports rather than food-processing exports and international tourism. The measurement of such a relationship is understated if the simultaneity between export expansion and G.D.P growth is ignored. [C 32, C 51, C 52, F 43]     

Testing for normality in linear regression models using regression and scale equivariant estimators

In this paper we provide a general solution to the problem of controlling the probability of a type I error in normality tests for the disturbances in linear regressions when using robust-regression residuals. We show that many classes of well-known robust regression estimators belong to the class of regression and scale equivariant estimators. It is these equivariance properties that are used to reduce the nuisance parameter space under the null, from which we develop Monte Carlo and Maximized Monte Carlo tests for the null of disturbance normality. Finally, we illustrate in a simulation experiment the potential power gains from using robust-regression residuals in testing this null hypothesis.     

New insights on earnings trends across skill groups and industries in West Germany

This paper provides an empirical analysis of the structure of earnings in West Germany across skill groups and industries. Our analysis is based on data from the German Socioeconomic Panel for the period 1984 to 1994. We estimate quantile regressions, both for the entire sample period and for each year separately, in order to obtain a finer picture of the earnings structure compared to conventional least squares methods. For robust standard error estimation, this study uses a block bootstrap procedure taking account of heteroskedasticity and autocorrelation in the error term. We also suggest a simple procedure to obtain a consistent estimate of inter-industry earnings variability. Our main findings are: first, pooled estimation comprising a uniform time trend is not rejected by the data, and second, the effects of human capital variables and industry dummies on earnings differ considerably across quantiles. First version: May 1998/Final version: April 2002 RID="*" ID="*"  We are grateful to an anonymous referee as well as to Thiess Büttner, Christian Dustmann, Wolfgang Franz, John Haisken-DeNew, Costas Meghir, Werner Smolny, Peter Winker, Volker Zimmermann, and seminar participants in Heidelberg, Kassel, Konstanz, and Paris for helpful comments. However, all errors are our sole responsibility. RID="*" ID="*"  We are grateful to an anonymous referee as well as to Thiess Büttner, Christian Dustmann, Wolfgang Franz, John Haisken-DeNew, Costas Meghir, Werner Smolny, Peter Winker, Volker Zimmermann, and seminar participants in Heidelberg, Kassel, Konstanz, and Paris for helpful comments. However, all errors are our sole responsibility.     

Spurious regressions driven by excessive volatility

This paper shows that the excessive volatility results in spurious regressions. The spuriousness can be driven by persistency in the error variances unlike the conventional spurious regressions that are generated by the persistency in the level of regression errors.     

Estimating the Heckscher–Ohlin model: Inverting the inverse matrix

This paper estimates the Heckscher–Ohlin model with annual US data from 1949 to 2006 for outputs of manufactures and services with inputs of fixed capital assets and the labor force. Difference equation and error correction regressions provide estimated coefficients for the comparative static system. Tariffs on manufactures primarily raise the capital return in the estimated Stolper–Samuelson results. Factor price equalization does not hold for labor and capital. Inverting the estimated system inverse matrix provides evidence on production. The suggestions are capital biased production of manufactures, strong substitution of capital for labor, and strong labor substitution in manufactures.     

Car ownership and use in Britain: a comparison of the empirical results of alternative cointegration estimation methods and forecasts

This paper addresses two problems faced by many forecasters in the transport sector, namely how to use a relatively small sample to forecast car ownership over a long period of time and avoid the difficulties caused by spurious or nonsense regressions. Five alternative estimation methods are used to test for cointegrating relationships between per capita car ownership (and use) and real per capita personable disposable income, real motoring costs and real bus fares. These are the Engle-Granger two-stage, the Phillips-Hansen fully modified, the Wickens-Breusch one-stage, the autoregressive distributed lag, and the Johansen maximum likelihood methods. The corresponding error correction models are estimated, and a comparison made between the derived short- and long-run demand elasticities for car ownership and use. The ex-post forecasting performance of the error correction models, together with an ARIMA model specification, is evaluated using a number of performance criteria. The long-range time series forecasts obtained from the cointegrating regressions are compared with those from the cross-sectional approach used by the UK Department of the Environment, Transport and the Regions, and the policy implications discussed.     

INVESTING IN INSTITUTIONS

Robust institutional change is difficult to achieve. However, it is more difficult for some countries than others. We use data on 69 countries between 1870 and 2000 to show that political instability does not always affect growth outcomes. We then develop a simple model to explain this fact in which the likelihood that “good” institutions are abandoned during periods of political uncertainty depends on the opportunity cost of doing so. We operationalize our model by using contract intensive money as a proxy for this initial investment in growth‐enhancing institutions. Cross‐sectional and panel growth regressions support the model's predictions.     

Does the labour supply of wives respond to husbands' wages? Canadian evidence from micro data and grouped data

Abstract. Using micro data and grouped data, we assess the extent to which Canadian wives adjusted their labour supply in response to changes in husbands' wages during the period 1980‐2000. Grouped data parameters based on weighted least squares and the unbiased‐error‐in‐variables estimator developed by Devereux (2004, 2007a,b) yield cross‐wage elasticities that are substantially higher (in absolute value) than those derived from OLS regressions run on micro data. Both grouping estimators indicate that the labour supply of Canadian wives responded strongly to changes in husbands' wages during the 1980s. For the 1990s, our estimates of wives' cross‐wage elasticity display greater dispersion.     

Rating Cooperative and Commercial Bank Bonds: a comparative approach

Abstract ** :  The primary objective of this article is to find whether bonds issued by commercial and cooperative banks are rated similarly or not. We then compare the performance of two quantitative methods, namely seemingly unrelated regressions (SURE) and recursive partitioning algorithm (RPA), at explaining bond ratings based on the same set of quantitative indicators. Using the regression model, cooperative banks' credit risk is more sensitive to the quality and size of assets. For commercial banks, elements relative to debt more clearly stand out. In the RPA model, a subtree for the financial cooperatives is created which provides evidence of some differentiation in the rating process. Also, the RPA model outperforms the parametric method whether performance is measured by the percentage of correct classification or the size of the average rating prediction error.     

Quantiles,corners, and the extensive margin of trade

We develop a simple method for the estimation of quantile regressions for corner solutions data (i.e., fully observed non-negative data that have a mixed distribution with a mass-point at zero), focussing particular attention on the case where the domain of the variate of interest is bounded both from below and from above. We use the proposed method to study the determinants of the extensive margin of trade and find that most regressors have very different impacts on different parts of the distribution.     

Estimating the causal effect of fertility on economic wellbeing: data requirements, identifying assumptions and estimation methods

This article aims to answer to what extent fertility has a causal effect on households’ economic wellbeing—an issue that has received considerable interest in development studies and policy analysis. However, only recently has this literature begun to give importance to adequate modelling for estimation of causal effects. We discuss several strategies for causal inference, stressing that their validity must be judged on the assumptions we can plausibly formulate in a given application, which in turn depends on the richness of available data. We contrast methods relying on the unconfoundedness assumption, which include regressions and propensity score matching, with instrumental variable methods. This discussion has a general importance, representing a set of guidelines that are useful for choosing an appropriate strategy of analysis. The discussion is valid for both cross-sectional or panel data.     

A Monte Carlo study of growth regressions

Using Monte Carlo simulations, this paper evaluates the bias properties of estimators commonly used to estimate growth regressions derived from the Solow model. We explicitly allow for measurement error, country-specific fixed effects and regressor endogeneity. An OLS estimator applied to a single cross-section of variables averaged over time (the between estimator) performs best in terms of the extent of bias on each of the estimated coefficients. Fixed-effects and the Arellano–Bond GMM estimator overstate the speed of convergence under a wide variety of assumptions, while the between estimator understates it. Finally, fixed effects and Arellano–Bond bias towards zero the slope estimates on the human and physical capital accumulation variables, while the between estimator and the Blundell–Bond system GMM estimator bias these coefficients upwards.      

Uncertainty and Consumer Durables Adjustment

We characterize infrequent durables stock adjustment by consumers who also derive utility from non-durable consumption flows in the presence of idiosyncratic income uncertainty. The data we analyse include subjective future income uncertainty measures, which we use as instruments in the estimation of relevant parameters of heterogeneous consumers' dynamic adjustment problems. The data feature two conceptually distinct sources of variation: cross-sectional heterogeneity of the sampled households' dynamic problems, and history-dependent heterogeneity in their situation during the observation period. We note that the latter should affect the likelihood but not the size of stock adjustment decisions, and find broad support for theoretical predictions in formal selection-controlled regressions based on this insight.