Nonlinear Modeling of Economic and Financial Time-Series: Volume 20

During the global financial crisis of 2008–2009, most developed and emerging economies and financial markets have recorded important financial losses. Those economies have experienced momentous corrections, and their assets were significantly devaluated, implying many losses and bankruptcies for banks, investors, and firms. Overall, despite continuing efforts made by governments and central banks to support their financial systems, most financial markets (stock markets, derivative markets, monetary markets, and currency markets) have been strongly affected by this crisis. Furthermore, the rapid transmission of the US subprime crisis to several European and Asian developed and emerging countries and the transformation into a global financial and economic crisis have revealed a high level of financial integration and linkage with the US market. The financial shocks have also induced negative feedbacks to macroeconomic indicators, suggesting significant relationships between financial markets and macroeconomies.

Purpose – The purpose of this chapter is to assess the role of collateralizable wealth and systemic risk in explaining future asset returns.

Methodology/approach – To test this hypothesis, the chapter uses the residuals of the trend relationship among housing wealth and labor income to predict both stock returns and government bond yields. Specifically, it shows that nonlinear deviations of housing wealth from its cointegrating relationship with labor income, hwy, forecast expected future returns.

Findings – Using data for a set of industrialized countries, the chapter finds that when the housing wealth-to-income ratio falls, investors demand a higher risk premium for stocks. As for government bond returns: (i) when they are seen as a component of asset wealth, investors react in the same manner and (ii) if, however, investors perceive the increase in government bond returns as signaling a future rise in taxes or a deterioration of public finances, then they interpret the fall in the housing wealth-to-income ratio as a fall in future bond premia. Finally, this work shows that the occurrence of crisis episodes amplifies the transmission of housing market shocks to financial markets.

Originality/value of chapter – These findings are novel. They also open new and challenging avenues for understanding the dynamics of the relationship between the housing sector, stock market and government bond developments, and the banking system.

Purpose – This chapter aims to investigate the stock market comovements between Mexico and the world capital market using nonlinear modeling tools.

Methodology/approach – We apply recent nonlinear cointegration and nonlinear error correction models (NECMs) to investigate the comovements between stock prices over the recent period.

Findings – While the previous literature only highlights some evidence of time-varying comovements, our chapter aims to specify the mechanism characterizing the comovement process through the comparison of two nonlinear error correction models (NECMs). It shows a nonlinear relationship between stock prices that are activated per regime.

Originality – Studying the integration hypothesis between stock markets over the recent financial crisis, our findings highlight strong evidence of significant comovements that explain the global collapse of stock markets in 2008–2009.

Purpose – The purpose of this chapter is to present an investigation on the dynamic linkages between global macro hedge funds and traditional financial assets of developed and emerging markets.

Methodology/approach – To explore relationships among these price indices, we analyse Granger causality and vector autoregression (VAR) dynamics through impulse response functions. Besides, multivariate cointegration is used to know long-term relationships between assets and allows risk-averse investors to reduce uncertainty. Finally, a vector error correction model (VECM) provides active asset managers the opportunity to anticipate short-term price movements.

Findings – Our results show that in a Granger causality sense, we observe long- and short-term relationships between global macro hedge funds and traditional financial assets for Canada, France and Germany. This implies that opportunities for international portfolio diversification are significantly lower for countries having relationships between assets. For Canada, France and Germany, the risk-averse investors can reduce their long-term volatility by investing according to the cointegrating vector, whereas active managers can benefit from the knowledge of short-term asset price movements. The VEC Pairwise Granger causality in the short term confirms our analysis of causality according to VAR models.

Originality/value of paper – These results are original because they help the investor to understand the dynamics of the relationship between global macro hedge funds and traditional financial assets.

Purpose – In this chapter, copula theory is used to model dependence structure between hedge fund returns series.

Methodology/approach – Goodness-of-fit tests, based on the Kendall's functions, are applied as selection criteria of the “best” copula. After estimating the parametric copula that best fits the used data, we apply previous results to construct the cumulative distribution functions of the equally weighted portfolios.

Findings – The empirical validation shows that copula clearly allows better estimation of portfolio returns including hedge funds. The three studied portfolios reject the assumption of multivariate normality of returns. The chosen structure is often of Student type when only indices are considered. In the case of portfolios composed by only hedge funds, the dependence structure is of Franck type.

Originality/value of the chapter – Introducing goodness-of-fit bootstrap method to validate the choice of the best structure of dependence is relevant for hedge fund portfolios. Copulas would be introduced to provide better estimations of performance measures.

Purpose – The aim of this chapter is to study the determinants of realignment expectations (as a measure of the exchange rate credibility).

Methodology/approach – To investigate this aspect, we apply a fixed effects model over the period 2001:01–2009:12 for a set of 14 European countries.

Findings – Using monthly data since the introduction of the euro, the chapter finds that standard macroeconomic phenomena and financial crisis over the selected period exerted a significant and positive impact on European realignment expectations. We also provide evidence that the 2008 global financial crisis has a significant effect on realignment expectations.

Originality/value of chapter – At our best knowledge, the single paper that studied a similar problem since the inception of the European Monetary System (EMS) in early 1979 was the research paper of Rose and Svensson (1993). Our findings are original in the sense that we find meaningful relationships between realignment expectations and financial crisis (systemic and nonsystemic crisis) and macroeconomic variables. Our research also wishes to contribute to the emergence of the recent studies on European exchange rate credibility.

Purpose – The purpose of this chapter is to investigate the linear and nonlinear short- and long-run relationships between the real price of oil and the US real effective exchange rate.

Methodology/approach – We use recent linear and nonlinear econometric techniques over the period 1973–2009.

Findings – Our main findings are that (i) there is significant evidence that both variables contain a unit root; (ii) the oil price and the US exchange rate are strongly linked in the short run; and finally (iii) there are some signs of nonlinearity in the oil–exchange rate relationship.

Originality – Using recent econometric techniques, we show that exchange rates are not a fundamental determinant of oil prices but exchange rate changes help to better forecast oil prices in the short run.

Purpose: Following the theoretical literature on growth model with externalities, the chapter aims at finding empirical evidence of the main sources of economic interdependencies in Europe.

Methodology/approach: A two-step econometric procedure is adopted. In the first step, in order to evaluate growth spillovers in Europe, cointegration relationships between indexes of industrial production per capita are estimated for 15 European countries. The estimated coefficients, interpreted as long-run elasticities between European countries, appear to be different between countries and unstable over time. In the second step, these coefficients are explained by trade, specialization, research and development (R&D), and macroeconomic variables.

Findings: Panel estimations show strong evidence in favor of a positive relationship between openness, country size, knowledge accumulation, and the long-run sensitivity to European income. European income spillovers are not explained by the specialization of trade and production. We conclude that countries that benefit the most from economic integration are the largest and those that invest the most in R&D.

Originality/value of chapter: The two-step approach adopted in this chapter is original and allows for measuring the impact of various determinants of externalities at the same time.

Purpose – The purpose of this chapter is twofold: to forecast gross domestic product (GDP) using nonparametric method, known as multivariate k-nearest neighbors method, and to provide asymptotic properties for this method.

Methodology/approach – We consider monthly and quarterly macroeconomic variables, and to match the quarterly GDP, we estimate the missing monthly economic variables using multivariate k-nearest neighbors method and parametric vector autoregressive (VAR) modeling. Then linking these monthly macroeconomic variables through the use of bridge equations, we can produce nowcasting and forecasting of GDP.

Findings – Using multivariate k-nearest neighbors method, we provide a forecast of the euro area monthly economic indicator and quarterly GDP, which is better than that obtained with a competitive linear VAR modeling. We also provide the asymptotic normality of this k-nearest neighbors regression estimator for dependent time series, as a confidence interval for point forecast in time series.

Originality/value of chapter – We provide a new theoretical result for nonparametric method and propose a novel methodology for forecasting using macroeconomic data.

Purpose – The aim of this chapter is to examine the constant proportion portfolio insurance (CPPI) method when the multiple is allowed to vary over.

Methodology/approach – A quantile approach is introduced under the dependent return hypothesis. We use for example ARCH-type models.

Findings – In this framework, we provide explicit values of the multiple as function of the past asset returns and other state variables. We show how the multiple can be chosen to satisfy the guarantee condition, at a given level of probability and for particular market conditions.

Originality/value of paper – We show in this chapter that it is possible to choose variable multiples for the CPPI method if quantile hedging is used and in the case of dependent log returns. Upper bounds can be calculated for each level of probability and according to state variables. This new multiple can be determined according to the distributions of the risky asset log return and volatility.