Marcel Aloy

Chercheur ,  Aix-Marseille Université , Faculté d'économie et de gestion (FEG)

Économétrie, finance et méthodes mathématiques
Aloy
Statut
Maître de conférences
Domaine(s) de recherche
Économétrie, Macroéconomie
Contact
marcel.aloy[at]univ-amu.fr
+33(0)4 13 94 20 02
Adresse

Maison de l'économie et de la gestion d'Aix
424 chemin du viaduc, CS80429
13097 Aix-en-Provence Cedex 2

Résumé Beta coefficients are the cornerstone of asset pricing theory in the CAPM and multiple factor models. This chapter proposes a review of different time series models used to estimate static and time-varying betas, and a comparison on real data. The analysis is performed on the USA and developed Europe REIT markets over the period 2009–2019 via a two-factor model. We evaluate the performance of the different techniques in terms of in-sample estimates as well as through an out-of-sample tracking exercise. Results show that dynamic models clearly outperform static models and that both the state space and autoregressive conditional beta models outperform the other methods.
Mots clés Autoregressive conditional beta, Dynamic conditional beta, State space, Multivariate GARCH, REITs, Real estate
Publication HAL Google Scholar
Résumé Alors que les politiques d'austérité et les réformes budgétaires décidées par la zone euro ont un impact de plus en plus perceptible sur les millions de citoyens européens, ce livre tente d'offrir une réponse aux questionnements qui traversent l'opinion publique. Au-delà d'un état des lieux de la zone euro, ce livre interroge la pertinence des choix d austérité et pose la question des ajustements à mettre en oeuvre dans une union monétaire entre pays hétérogènes.
HAL Google Scholar
Résumé Estimation methods of bivariate fractional cointegration models are numerous and have in most cases non-equivalent asymptotic and finite sample properties, implying difficulties in determining an optimal estimation strategy. This paper addresses this issue by means of simulations and provides useful guidance to practitioners. Our Monte Carlo study reveals the superiority of techniques that estimate jointly all parameters of interest, over those operating in two steps. To illustrate the empirical relevance of our results, we propose a co-persistence analysis of two stock market realized volatility series.
Mots clés Economie quantitative
Publication HAL Google Scholar
Mots clés Economie quantitative
HAL Google Scholar
Résumé This paper attempts to provide evidence of “shift-volatility” transmission in the East Asian equity markets. By “shift-volatility”, we mean the volatility shifts from a low level to a high level corresponding respectively to tranquil and crisis periods. We examine the interdependence of equity volatilities between Hong-Kong, Indonesia, Japan, Malaysia, the Philippines, Singapore, Thailand and the United States. Our main issue is whether shift-volatility needs to be considered as a regional phenomenon, or from a more global perspective. We propose several indicators that are be useful to guide the investors in their arbitrage behavior in the different regimes: the duration of each state, the sensitivity of the volatility in a market following a change in the volatility in another market. Finally, we are able to identify which market can be considered as leading markets in terms of volatility.
Mots clés Finance general, Macroeconomics, Monetary Economics, Financial Economics, Quantitative Finance
Publication HAL Google Scholar
Résumé This article contributes to the recent empirical literature on financial repression and focuses on the French case since the end of World War II. We find that the fiscal adjustment needed to lower the debt ratio has been smaller during the years of financial repression in comparison with those of liberalized financial markets. This was possible because the real interest rates were low. We conduct a counterfactual analysis to see whether the vulnerability of public finances would have been different, if, since the late 1980s, the governments had continued carrying out the same financial repression policies. We answer affirmatively showing that the cost of debt service would have been reduced.
Mots clés Economie quantitative
Publication HAL Google Scholar
Résumé This paper proposes a new fractional model with a time-varying long-memory parameter. The latter evolves nonlinearly according to a transition variable through a logistic function. We present an LR-Based test that allows to discriminate between the standard fractional model and our model. We further apply a nonlinear least squares estimation method to estimate the long-memory parameter. We present an application to the unemployment rate in the United States from 1948 to 2012.
Mots clés Economie quantitative
Publication HAL Google Scholar
Résumé The recent empirical literature supports the view that most of the international stock prices are not pairwise cointegrated. However, by using fractional cointegration techniques, this paper shows that France, Germany, Hong Kong, and Japan stock prices indices are pairwise fractionally cointegrated with US stock prices. Equilibrium errors are mean reverting with half-life lying between 2 and 12 days. It is worthwhile noting that emerging markets like Brazil and Argentina are not pairwise cointegrated with the US stock market. These new results have important implications for asset pricing and international portfolio strategy.
Mots clés Social Sciences &amp, Humanities
Publication HAL Google Scholar
Résumé Estimation methods of bivariate fractional cointegration models are numerous. In most cases they have non-equivalent asymptotic and finite sample properties, implying diffculties in determining an optimal estimation strategy. In this paper, we address this issue by means of simulations and provide useful guidance to practitioners. Our Monte Carlo study reveals the superiority of techniques that estimate jointly all parameters of interest, over those operating in two steps. In some cases, it also shows that estimators originally designed for the stationary cointegration, have good finite sample properties in non-stationary regions of the parameter space.
Mots clés Fractional cointegration, Monte Carlo simulation, Whittle estimation, Frequency analysis
Télécharger HAL Google Scholar
Résumé This paper attempts to provide evidence of "shift-volatility" transmission in the East Asian equity markets. By shift-volatility, we mean the volatility shifts from a low level to a high level, corresponding respectively to tranquil and crisis periods. We examine the interdependence of equity volatilities between Hong-Kong, Indonesia, Japan, Malaysia, the Philippines, Singapore, Thailand and the United States. Our main issue is whether shift-volatility needs to be considered as a regional phenomenon, or from a more global perspective. We find that the timing/spans of high volatility regimes correspond adequately to years historically documented as those of crises (the Asian crisis and the years following the 2008 crisis). Moreover, we suggest different indicators that could be useful to guide the investors in their arbitrage behavior in the different regimes: the duration of each state, the sensitivity of the volatility in a market following a change in the volatility in another market. Finally, we are able to identify which market can be considered as leading markets in terms of volatility.
Mots clés Regime shifts, Equity volatility, East Asia, TVPMS
Télécharger HAL Google Scholar
Résumé This paper proposes a new fractional model with a time-varying long-memory parameter. The latter evolves nonlinearly according to a transition variable through a logistic function. We present a LR-based test that allows to discriminate between the standard fractional model and our model. We further apply the nonlinear least squares method to estimate the long memory parameter. We present an application to the unemployment rate in the United -States from 1948 to 2012.
Mots clés Time varying parameter, Nonlinearity, Long-memory, Logistic
Télécharger HAL Google Scholar
Résumé Estimation of bivariate fractionally cointegrated models usually operates in two steps: the first step is to estimate the long run coefficient (\beta) whereas the second step estimates the long memory parameter (d) of the cointegrating residuals. We suggest an adaptation of the maximum likelihood estimator of Hualde and Robinson (2007) to estimate jointly \beta and d, and possibly other nuisance parameters, for a wide range of integration orders when regressors are I(1). The finite sample properties of this estimator are compared with various popular estimation methods of parameters \beta (LSE, ADL, DOLS, FMLS, GLS, MLE, NBLS, FMNBLS), and d (LPE,LWE,LPM,FML) through a Monte Carlo experiment. We also investigate the crucial question of testing for fractional cointegration (that is, d < 1). The simulation results suggest that the one-step methodology generally outperforms others methods, both in terms of estimation precision and reliability of statistical inferences. Finally we apply this methodology by studying the long-run relationship between stock prices and dividends in the US case.
Mots clés Fractional cointegration, Long memory, Monte Carlo Experiment, Cointegration test
Télécharger HAL Google Scholar