دانلود مقاله ISI انگلیسی شماره 12543
ترجمه فارسی عنوان مقاله

خطر بازگشت مبادله در بازار سرمایه حوضه اقیانوس آرام

عنوان انگلیسی
Risk-Return Trade-Off in the Pacific Basin Equity Markets
کد مقاله سال انتشار تعداد صفحات مقاله انگلیسی
12543 2014 18 صفحه PDF
منبع

Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)

Journal : Emerging Markets Review, Available online 30 January 2014

ترجمه کلمات کلیدی
توزیع - واریانس شرطی - چولگی شرطی - موقتی - نمونه داده های مخلوط - بازارهای سرمایه در حوزه اقیانوس آرام - خطر بازگشت تجارت کردن
کلمات کلیدی انگلیسی
Binormal distribution, Conditional variance, Conditional skewness, Intertemporal CAPM, Mixed data sampling, Pacific basin equity markets, Risk-return trade-off,
پیش نمایش مقاله
پیش نمایش مقاله  خطر بازگشت مبادله در بازار سرمایه حوضه اقیانوس آرام

چکیده انگلیسی

We conduct an empirical study of risk-return trade-off in fourteen Pacific basin equity markets using several volatility estimators, including five variants of GARCH class, equally weighted rolling window volatility, and mixed data sampling (MIDAS), as well as binormal GARCH (BiN-GARCH) model which allows for non-zero conditional skewness in returns. Our findings imply that the BiN-GARCH model, which allows for time-variation in the conditional skewness and market price of risk, captures the expected positive risk-return relationship in eleven out of fourteen markets studied. In comparison, symmetric skewness models such as MIDAS or GARCH variants fail to capture positive and statistically significant market price of risk estimates. These results provide support for the growing literature on the necessity of modeling conditional higher moments in financial research.

مقدمه انگلیسی

In the past twenty years, many studies in empirical asset pricing have devoted considerable energy to testing the systematic trade-off between expected returns and risk, characterized by [31] intertemporal capital asset pricing (ICAPM). The majority of empirical studies postulate a positive and linear relation between the conditional variance and expected excess market returns, with time-invariant market price of risk. Moreover, the majority of studies assume symmetry in the conditional distribution of excess returns. These assumptions practically translate into a race to come up with the best model for conditional volatility. However, there is no consensus on even the most basic theoretical properties of the risk-return relationship; see [33]. In this study, we show that the problem is not the volatility model used, but failure to account for conditional higher moments and their role in the risk-return relationship. We use several variants of existing empirical methodologies to test the risk-return relationship in 14 Pacific rim financial markets. We find that the model that does not require symmetry in the conditional distribution of excess returns outperforms those that require this restriction. We find empirical evidence supporting a positive risk-return relationship in 13 out of 14 markets studied, using binormal GARCH (henceforth, BiN-GARCH) methodology of [11]. Other methods, including rolling estimator of [13], various GARCH specifications, or mixed data sampling (henceforth, MIDAS) of [16], [17] and [18], deliver significantly weaker results. We believe that our results provide significant empirical support for the idea that to build reliable risk management measures in equity markets in general and Pacific basin markets in particular, one needs to consider modeling conditional higher moments such as conditional skewness, in addition to the traditional modeling of the first two conditional moments. This paper contributes to the existing literature in two important directions. First, it provides an empirical assessment for the ability of several traditional and more recent econometric methodologies in gauging the elusive risk-return trade-off relationship, in the context of non-U.S. markets. Second, this paper empirically documents the existence of asymmetry in conditional distribution of returns, non-linearity in the risk-return trade-off relation, and time-variation in the conditional skewness across the most important region in the world economy in the 21st century.1 In general, empirical evidence on risk-return trade-off is quite inconclusive. [6] and [32], and more recently [5], find a significantly negative conditional relationship. [23] and [20] find both a positive and a negative relation depending on the method used. On the other hand, [13], [1] and [7] find a positive but mostly insignificant relation between the conditional variance and the conditional expected returns. [16] and [29] find a positive and significant relationship in the U.S. data. In particular, [16] report success in capturing a time-invariant, positive, and statistically significant market price of risk for monthly market returns data from 1928 to 2000. They use daily data for the same period and mixed data sampling (MIDAS) regression methodology to perform statistical assessment of ICAPM. More recently, [33] use regression trees to show that the risk-return relationship is state dependent and nonlinear. [11] show that the risk-return relationship is nonlinear, time-varying, and crucially depends on the dynamics of conditional skewness of returns. They show that inclusion of conditional skewness in estimation delivers robust, positive risk-return tradeoff in S&P500 and international data. teGhyselsPlazziValkanov10 show that incorporation of measures of conditional skewness is vital in modeling the dynamics of equity returns, particularly for emerging markets, such as the majority of markets studied in our paper. The rest of the paper is organized as follows. In Section 2, we introduce the data used in this paper. Section 3 introduces the various methodologies used in the paper and discusses the empirical results. Section 4 concludes.

نتیجه گیری انگلیسی

We study risk-return tradeoff relationship in fourteen Pacific basin financial markets. The main question addressed in the paper is whether modeling, and pricing the market price of risk, for the first two moments is adequate for capturing the dynamics of risk-return tradeoff relationship. We empirically demonstrate that in addition to conditional mean and volatility, one needs to model the conditional skewness. Thus, we provide additional support for the relatively new and growing literature on the importance of modeling of conditional skewness, such as [24], [25], [15] and [11]. We show that regardless of the returns volatility model used, assuming zero conditional skewness which means symmetry in returns, leads to failure in capturing the expected positive market price of risk. We also empirically demonstrate that by allowing for time-varying and non-zero conditional skewness in BiN-GARCH model, we successfully capture a positive market price of risk in all but one of the markets studied. These findings have an important impact on the way we study risk-reward relationship. We show that market price of risk modeled as a time-varying process, and not a constant parameter, is better supported by the data in Pacific basin markets. This implies that the level of effective risk tolerance, following [31] formulation, is time-varying. In turn, this observation has profound implications for financial activities such as portfolio choice and option pricing that make implicit or explicit assumptions regarding risk tolerance of market participants. In fact, [8] and [12] emphasize the importance of modeling conditional skewness in improving option pricing and portfolio selection. As [11] demonstrate, a severe negative shock to the financial system translates into an increase downside semivariance -hence an increase in relative downside volatility - and an asymmetric increase in conditional skewness characterized by Pearson mode skewness. In turn, these outcomes translate into an increase in market price of risk. These observations are substantiated by what we find in Pacific basin markets. On the other hand, we show that ignoring conditional skewness leads to considerable misspecification in the econometric models of risk. Thus, to construct more accurate risk management tools such as Value-at-Risk (VaR) or expected shortfall (ES), a careful modeling of conditional higher moments, such as conditional skewness, is crucial.