Résumés des communications

Multivariate Time Series Analysis: Theory and Methods / Applications in Economics and Finance

Abstract:

Multivariate time series are widely used in economics since a substantial part of economic theory generally deals with long-run equilibrium relationships generated by market forces and behavioral rules. Such multivariate models arise when one simultaneously observe multiple time series instead of a single univariate time series. In the study of multivariate process, a framework is needed to describe not only the different properties of each component of the process individually, but also links that may exist between the various components. To analyze and modelize a multivariate time series is basically to understand the dynamic relationships over time between different series and improve the accuracy of predictions for each series using not only the information in the past for each one but also the information in the history of the other series.

In this lecture, we will focus mainly on the recent developments in multivariate time series processes and their applications. Often, the first step in constructing a model for a specific purpose or for a particular sector of an economy is to decide on the variables to be included in the analysis. At this stage it is usually important to take into account what economic theory has to say about the relations between the variables of interest. Suppose we want to analyze the transmission mechanism of monetary policy. An important relation in that context is the money demand function, which describes the link between the real and the monetary sector of the economy. In this relationship a money stock variable depends on the transactions volume and opportunity costs for holding money. As an example, Lutkepohl (2004) considered German M3 as the money stock variable, GNP as a proxy for the transactions volume, a long-term interest rate R as an opportunity cost variable, and the inflation rate Dp, where p denotes the log of the GNP deflator. The latter variable may be regarded as a proxy for expected inflation, which may also be considered an opportunity cost variable.

First, we will introduce the Vector Autoregressive processes (VAR). These processes are a suitable model class for describing the data generation process (DGP) of a small or moderate set of time series variables. In these models all variables are often treated as being a priori endogenous, and allowance is made for rich dynamics. Restrictions are usually imposed with statistical techniques instead of prior beliefs based on uncertain theoretical considerations. Furthermore, special interest occurs if several variables are driven by a common stochastic trend. In that case they have a particularly strong link that may also be of interest from an economic point of view. These variables are called Cointegrated. This concept was introduced by Engle-Granger (1987) and used in many recent studies across different fields. Henceforth, if cointegrating relations are present in a system of variables, the VAR form is not the most convenient model setup. In that case it is useful to consider specific parameterizations that support the analysis of the cointegration structure.

The resulting models are known as vector error correction models (VECMs) or vector equilibrium correction models. Many other types of VAR models will be also discussed with different applications in economics and finance.

Keywords: Vector Autoregressive Models, Vector Error Correction Models, Cointegrated models, Causality and Independence tests, Multivariate GARCH models, Monetary policy.

Return and Volatility Spillovers in the Moroccan Stock Market During The Financial Crisis

Abstract:

The aim of this paper is to investigate the return and volatility linkages among Moroccan stock market with that of U.S. and three European countries (France, Germany and U.K.) before and during the financial crisis. More specifically, we use stock returns in MASI, CAC, DAX, FTSE and NASDAQ as representatives of Moroccan, French, German, British and U.S. markets respectively. The data sample frequency is daily and spans from January 2002 to December 2012 excluding holidays. Using the estimation results of bivariate VAR-BEKK GARCH model, we analyze the return and volatility spillover effects between the Moroccan market and the other considered markets. Moreover, the identification of break point due to the subprime crisis is made by Lee-Strazicich (2003,2004) and Bai-Perron (1998, 2003) structural break tests. The empirical findings provide clear evidence of stronger linkages between the Moroccan market and the four other considered stock markets have been created during the subprime financial crisis period.

Keywords: Return and volatility spillovers; multivariate GARCH model; financial crisis; stock markets; break identification; conditional correlation.

Oil Supply and Demand Shocks and Stock Price: Evidence for some OECD Countries

Abstract:

This paper examines the interactive relationships between oil price shocks and stock market in 11 OECD countries using Vector Error Correction Models (VECM). Considering both world oil production and world oil prices to supervise for oil supply and oil demand shocks, strong evidence of sensitivity of stock market returns to the oil price shocks specifications is found. As for impulse response functions, it is found that the impact of oil price shocks substantially differs along the different countries and that the results also differ along the various oil shock specifications. Our finding suggests that oil supply shocks have a negative effect on stock market returns in the net oil importing OECD countries. However, the stock market returns are negatively impacted by oil demand shocks in the oil importing OECD countries, and positively impacted in the oil exporting OECD countries.

Keywords: Oil price; Stock market return; Oil supply shocks; Oil demand shocks, Vector Error Correction Models.

Financial market contagion during the global financial crisis: evidence from the Moroccan stock market.

Abstract:

In this paper, we aim at the study of the contagion of the global financial crisis (2007–2009) on Moroccan stock market. Our study focuses to examine whether contagion effects exist on Moroccan stock market, during the current financial crisis. Following Forbes and Rigobon (2002), we define contagion as a positive shift in the degree of comovement between asset returns. We use stock returns in MASI, CAC, DAX, FTSE and NASDAQ as representatives of Moroccan, French, German, British and US markets, respectively. To measure the degree of volatility comovement, time-varying correlation coefficients are estimated by flexible dynamic conditional correlation (DCC) multivariate GARCH model. We investigate empirical studies using the DCC-GARCH framework to test the contagion hypothesis from US and European markets to the Moroccan one.

Keywords: multivariate GARCH model; financial crisis; contagion hypothesis; break identification; conditional volatility; volatility comovement; financial markets; derivatives.

Exact Quasi-Optimal and Adaptive Inference in Linear and Nonlinear Regression Models

Abstract:

We propose point-optimal sign-based tests for linear and nonlinear regression models with dependent data. The tests are exact, distribution-free, and robust against heteroskedasticity of unknown form. We also propose an adaptive approach based on split-sample technique to choose an alternative at which the power curve is close to the power envelope. These applied to test for the predictability of stock returns at different horizons.

Keywords: sign-based test; dependent data; point-optimal test; nonlinear models; heteroskedasticity; exact inference; distribution free; power envelope; sample-split; adaptive approach; projection.