Publications

Abstract On the face of it, econometrics and machine learning share a common goal: to build a predictive model, for a variable of interest, using explanatory variables (or features). However, the two fields have developed in parallel, thus creating two different cultures. Econometrics set out to build probabilistic models designed to describe economic phenomena, while machine learning uses algorithms capable of learning from their mistakes, generally for classification purposes (sounds, images, etc.). Yet in recent years, learning models have been found to be more effective than traditional econometric methods (the price to pay being lower explanatory power) and are, above all, capable of handling much larger datasets. Given this, econometricians need to understand what the two cultures are, what differentiates them and, above all, what they have in common in order to draw on tools developed by the statistical learning community with a view to incorporating them into econometric models.
Keywords Least squares, Modelling, Econometrics, Big data, Learning
Publication HAL Google Scholar
Abstract The article investigates the effects of Employment Protection Legislation (EPL) on capital and skills according to the intensity of international competition. Grounded on a panel data sample for 14 OECD countries and 18 industries from 1988 to 2007, and a difference-in-difference approach, we find that strengthening EPL: (i) leads to a capital-labour substitution in favour of non ICT non R&D capital to the detriment of employment, this effect being mitigated in industries highly exposed to international competition; (ii) lowers ICT capital and, even more severely, R&D capital relatively to other capital components; and (iii) works at the relative disadvantage of low-skilled workers. Strengthening EPL can therefore be an impediment to organizational and so technological change and risk taking on globalized markets. An illustrative simulation suggests that structural reforms weakening EPL could have a significant favorable impact on firms’ ICT and R&D investment and on hiring low-skilled workers.
Publication HAL Google Scholar
Abstract We evaluate the introduction of various forms of antihypertensive treatments in France with a distribution-sensitive cost-benefit analysis. Compared to traditional cost-benefit analysis, we implement distributional weighting based on equivalent incomes, a new concept of individual well-being that does respect individual preferences but is not subjectively welfarist. Individual preferences are estimated on the basis of a contingent valuation question, introduced into a representative survey of the French population. Compared to traditional cost-effectiveness analysis in health technology assessment, we show that it is feasible to go beyond a narrow evaluation of health outcomes while still fully exploiting the sophistication of medical information. Sensitivity analysis illustrates the relevancy of this richer welfare framework, the importance of the distinction between an ex ante and an ex post approach, and the need to consider distributional effects in a broader institutional setting.
Keywords Equivalent income, Distributional weights, Cost-benefit analysis, Antihypertensive treatment
Publication HAL Google Scholar
Abstract Constraints that prevent women from working longer hours are argued to be important drivers of the gender wage gap in the United States. We provide evidence that in couples where the wife's working hours exceed the husband's, the wife reports lower life satisfaction. By contrast, there is no effect on the husband's satisfaction. The results still hold when controlling for relative income. We argue that these patterns are best explained by perceived fairness of the division of household labor, which induces an aversion to a situation where the wife works more at home and on the labor market.
Publication HAL Google Scholar
Abstract This paper shows that a large dimensional vector autoregressive model (VAR) of finite order can generate fractional integration in the marginalized univariate series. We derive high-level assumptions under which the final equation representation of a VAR(1) leads to univariate fractional white noises and verify the validity of these assumptions for two specific models.
Keywords Marginalization, Long memory, Final equation representation, Vector autoregressive model
Publication HAL Google Scholar
Abstract This paper deals with the effects of economic integration in a 2x 2x 2 model of overlapping generations. We distinguish between a non-tradable and a tradable sector which use human and physical capital. We show that the preference for non-tradable consumption in total consumption expenditure and sectoral productivities are crucial factors to determine which country does benefit from integration in terms of economic growth. Short-run and long-run effects of integration may differ, especially when countries are heterogeneous and when there exist high cross border externalities in education. Moreover, an impatient country may lose to integration when it has a comparative advantage in the tradable sector and/or when the preference for non-tradable goods is high.
Publication HAL Google Scholar
Abstract This articles focuses on the recent research efforts to incorporate income, wage and wealth inequality in macroeconomic models. I start by reviewing recent models on the impact of inequality on, on the one hand, long-run growth and, on the other, and macroeconomic fluctuations. The articles then reviews the literature concerned with the macroeconomic determinants of wage and wealth inequality. It concludes by discussing a number of possible avenues of research that seem to me particularly important, such as the impact of macroeconomic policy on distribution or the effect that firm size can have on both growth and wage inequality.
Publication HAL Google Scholar
Keywords Tax competition, Top income earners, Migration
Publication HAL Google Scholar
Abstract In this study, we model realized volatility constructed from intra-day highfrequency data. We explore the possibility of confusing long memory and structural breaks in the realized volatility of the following spot exchange rates: EUR/USD, EUR/JPY, EUR/CHF, EUR/GBP, and EUR/AUD. The results show evidence for the presence of long memory in the exchange rates' realized volatility. FromtheBai-Perrontest,wefoundstructuralbreakpointsthatmatch significant events in financial markets. Furthermore, the findings provide strong evidence in favour of the presence of long memory.
HAL Google Scholar
Keywords Customer discrimination Matching frictions Neighborhood externalities Housing market
Publication HAL Google Scholar