Pei-Fen Chen

Asymmetric effects of investor sentiment on industry stock returns: Panel data evidence

Abstract This article employs a state-of-the-art panel threshold model by allowing for regime intercepts, in order to shed new light on the asymmetric/nonlinear effects of local and global sentiments on expected industry stock returns among 11 Asian countries during the period from 1996 to 2010. Empirical evidence demonstrates that once the regime intercept is included, the asymmetric effects of global sentiment on oil & gas, financials, and health care industry returns become less under optimism, as compared with under pessimism. More critically, the positive (negative) impact of global sentiment above (under) the threshold turns significant, indicating that global optimism leads industry returns to be overvalued, while pessimism leads them to be undervalued. For local market sentiment, our results support that higher local sentiment enhances the returns of basic materials, telecommunications, and utilities industries. The empirical results confirm that the nexus of industry returns and investor sentiments is subject to change between different sentimental intervals.

Do CO2 emission levels converge among 21 OECD countries? New evidence from unit root structural break tests

This study empirically re-investigated whether carbon dioxide (CO2) emissions series were stationary in 21 OECD countries during the 1960–2000 period. A suite of test statistics were employed, proposed by Sen (2003), with a model that simultaneously allows for a break within the context of Perrons (1989) mixed intercept and slope (Model C). The distinction between this study and previous ones lies in its control for breaks. Compared with the results from traditional unit root tests, the empirical findings provide further evidence that relative per capita CO2 emissions were stationary and were stochastically converging. In addition, structural breaks are identified in each country, and some important policy implications emerge from the results.

Money demand function versus monetary integration: Revisiting panel cointegration among GCC countries

This paper applies the newly developed panel cointegrated technique [R. Larsson, J. Lyhagen, M. Lothgren, Likelihood-based cointegration tests in heterogeneous panels, Econom. J. 4 (1) (2001) 109-142] that allows for multiple cointegrated relationships to empirically re-examine the long-run money demand function for six selected countries of the Gulf Cooperation Council (GCC) in 1979-2000. At the center of focus is a discussion on the regional money demand phenomena with challenges and potential benefits, because of these GCC member adoption this monetary integration policy. It is determined that there are at least two cointegrated relations in the four-dimensional vector error correction model for the variables of the real money balance, the real scale variable, the nominal interest rate, and the exchange rate. The coefficient restriction test is also conducted, and it substantiates that the full panel test significantly rejects the hypothesis of the quantity theory of money for the long-run elasticity of income equal to unity, and we are able to reject the null hypothesis when the semi-elasticities of the nominal interest rate and exchange rate are equal to zero. Some critical policy implications emerge from the results.

Testing linearity in a cointegrating STR model for the money demand function: International evidence from G-7 countries

The motivation behind this paper is to re-investigate the stability of the long-run money demand function (MDF) in a non-linear cointegrating framework for G-7 countries. Previous studies on non-linearity in the MDF are only related to the short-run dynamics and assume that long-run cointegrating relations are linear, which according to economic theory need not be the case. Thus, we really need to focus on the variables in the long-run MDF and their determinants through the adoption of a cointegrating smooth transition regression (CSTR) test developed by [I. Choi, P. Saikkonen, Testing linearity in cointegrating smooth transition regressions, Economet. J. 7 (2004) 341-365]. The reason is due to this model being more general than the traditional STR model in that it may contain several transition functions and has more than a single transition variable. Our evidence demonstrates the existence of a non-linear cointegrating relationship, and as such several transition variables should be of more concern under the non-linear hypothesis. Overall, we propose more possibilities that will bring about the unstable phenomenon of the long-run MDF.

Is Rate of Stock Returns a Leading Indicator of Output Growth? In the Case of Four East Asian Countries

The link between stock returns and economic growth has been an important research topic in the financial economic literature. The purpose of this study is to employ a threshold vector autoregressive (TVAR) approach in order to investigate the non-linear relationship between stock returns and output growth in the four East Asian countries: Taiwan, Japan, Korea, and Malaysia. The causality between the two under different stock returns regimes are also examined in this study. The empirical findings of this modest study indicate that there is an asymmetric relationship between stock returns and industrial production growth in the four East Asian countries. Finally, the findings from the generalized impulse response function of the VAR model reveal that each country’s stock market disturbance has a greater impulse effect on output growth under bad news regime than under good news regime.