Analysis of Foreign Exchange Spot Markets - Research Proposal Example

Abstract This paper is an attempt to dwell on the issue of information and volatility links across financial markets. These variables are contributed by the model proposed by Fleming, Kirby & Ostdek (1998), who have posited that economic or financial connections between two markets are fashioned by the common information that these markets share and the cross-hedging activity or the position that is taken in one market to hedge risk in a position taken in another market. The period that coincided with the global financial crisis in 2007 until 2009 was specifically identified as source of FX rates for analysis because instances of crises render the financial markets more volatile than in usual circumstances. It would have been a full blown subscription to the generalized method of moments (GMM); but, in the course of its statistical treatment of pertinent data, this paper has found out that there seems to be an absence of a strong correlation between the Euro and Yuan FX rates. To account for this, a brief notation on how Euro and Yuan historically fared during the crisis was provided. Introduction This paper is particularly about information and volatility links across foreign exchange (FX) spot markets (see Treepongkaruna & Gray, 2009), especially during the instance of global financial crisis in 2007 until 2009. At the start of studies relative to volatility links, the focus by scholars was on equity and other related derivative markets in the US (see Fleming, Kirby & Ostdek, 1998). Recently, it was extended to money, bond, and other derivatives in and out of America. Fleischer (2003) has adapted the studies of Fleming, Kirby & Ostdek (1998) to the markets in Australia. Antell (2004) examines the volatility links in Finish stock, bond and money markets. Christiansen (2004) has done a study on volatility spillover from the US and the European markets. Cheong, Nor & Isa (2007) investigate the volatility causal linkages among the eight major sectorial indices in Malaysian stock market. Cai & Treepongkaruna (n.d.) did more than study either the VRV dynamics in a financial market, or the VRV dynamics among financial markets. Rather, they examined the former in a cross market context and scrutinized the latter as the intra-market VRV dynamics are under control. Similarly, the expansion of studies on volatility links has involved the use of different tools. The traditional measure of volatility was variance or standard deviation, which is unconditional and does not recognize the presence of interesting patterns in asset volatility (Olowe, 2009). Engle (1982) introduced the autoregressive conditional heteroskedasticity (ARCH) to model volatility. Bollerslev (1986) endorsed generalized autoregressive conditional heteroskedasticity (GARCH). Fleming, Kirby & Ostdek (1998) and Fleischer (2003) estimated cross market linkages with the use of Generalized Method of Moments (GMM), which they have jointly found to be better than the other commonly used proxies for volatility. Antell (2004) likewise availed of the GMM and the vector-autoregressive EGARCH framework or Exponential GARCH model (see Nelson, 1991) in his study of the Finish financial market. Ogum (2010) applies the dynamic bivariate multivariate GJR-GARCH model to daily stock return data of twelve emerging market economies and the US during the period that encompassed the subprime crisis in order to investigate the volatility spillovers from the US to other economies and test for changes in the transmission mechanism – i.e., contagion – during the global finance crisis. Other derivatives of the GARCH model include the IGARCH (Engle & Kroner, 1995), GARCH-in-Mean (GARCH-M) (see Engle, Lilien & Robins, 1987), the standard deviation GARCH model (Taylor, 1986; Schwert, 1989), TARCH or Threshold ARCH and Threshold GARCH (Zakoian, 1994; Glosten, Jaganathan & Runkie, 1993), and Power ARCH model (Ding, Granger & Engle, 1993). Cai & Treepongkaruna (n.d.) applied the Structural Vector Autoregressive (SVAR) model to study the cross market VRV dynamics among Deutsch Mark (DM), Japanese Yen (JPY) and Indonesia Rupiah (IDR) in the FX spot market during the 1997 Asian financial crisis. As market volatility has been a source of deep concern to investors, analysts, brokers and regulators (Olowe, 2009), these studies as a matter of fact serve as basis for investment diversification strategies, hedging strategies, and regulatory policies across the financial markets (Cheong, Nor & Isa, 2007). Volatility links between currencies, in particular, affect the movement of FX rates and, thus, have strong impact on FX trading and risk management activities. For one, cross-currency hedging can only be effective when the trader properly understands the nature of volatility linkages between currencies. Volatility affects the pricing of derivatives, too. Thus, accounting for volatility linkages allows for more accurate pricing of FX derivatives. Likewise, cross-currency correlations of volatility determine the net volatility of those who have exposure to more than one currency. As a matter of consequence, then, properly specified information and volatility links may be incorporated into risk measurement and management systems and in setting aggregate position limits across trading desks. Volatility links are useful for policy making purposes. Central banks’ decisions in fact affect the over-all market sentiment and risk management systems (see Treepongkaruna & Gray, 2009). And, these studies serve their purpose even better on occasions of financial turmoil when the financial markets are more volatile than usual. For, volatility in the financial market – or the variability of stock prices changed – is taken as barometer of the vulnerability of financial markets (Olowe, 2009). With increased volatility, financial markets afford greater opportunities while at the same time necessitate greater care and prudence in decision making. With figures that are from the Federal Reserve Statistical Release (FRSR, 2009) and OANDA (2010), this paper would want to examine information and volatility links across financial markets. These data culled from FX market are rich set of data that would facilitate an examination of the information and volatility links. These involve the average daily FX turnover of $3.21 billion in April, 2007 (BIS, 2007), and its approximately twenty per cent (20%) growth to $4 billion in April 2010 (Lien, 2010). Following the model originally proposed by Tauchen & Pitts (1983) and extended by Fleming, Kirby & Ostdek (1998), this paper would take into account the two sources of information and volatility links – namely, the information that simultaneously affect the expectations in more than one market, such as the decision of the US Federal Reserve pertinent to interest rate or monetary policy (see the studies on the dynamics of volume-return-volatility [VRV] by Nguyen & Daigler, 2006; Kocagil & Schachmurove, 1998), and the information spill over that is occasioned by cross-market hedging (see, for instance, the studies by Treepongkaruna & Gray, 2009; Fleischer, 2003; Fung & Patterson, 1999). It is going to apply GMM to estimate the trading model for the three currencies of US dollars (USD), Euro (Euro), and Chinese Yuan (CNY). Methodology The data used in this paper is culled from the US Federal Reserve Statistical Release (FRSR, 2010) for the following currencies: USD, Euro and CNY. Needing to adopt a numeraire as all papers doing this kind of study do (see Treepongkaruna & Gray, 2009), this paper has adopted as if each of the Euro and CNY were an asset available to an American investor. The (historical) data were from June 2007 to May 2009, covering the period of the 2007-2009 financial crises that began in the US and afflicted the entire world. During this period of crises, financial markets volatility was higher than during other periods. Dacorogna et al. (1993) has come up with a formula to determine the daily FX returns. This serves the idea by Goodhart, Ito & Payne (1996) that hourly computed FX returns are good estimation of the true underlying returns. However, OANDA (2010) and FRSR (2009) provide as historical data the daily average of FX returns (see Appendix A). Hence, in this paper, there is no more need to take into account the last bid and the ask FX rates each hour just to form the return series for the USD, Euro and CNY. Or, it has become dispensable to make several hundreds of observations for every currency (as done by Treepongkaruno & Gray, 2009). What is important is that with the data from OANDA (2010) and FRSR (2009) the impact of the pattern of trading throughout the day on the currency prices and the estimates of volatility spillover are still realizable. After the average FX rates have been ascertained, the GMM estimation begins. The GMM estimates the parameter vector by minimizing the sum of squares of the differences between the population moments and the sample moments, with the use of variance of the moments as a metric (Hall, 1999). In Treepongkaruno & Gray (2009), there is a general following of the formula by Fleming, Kirby & Ostdek (1998). Result Descriptive statistics, at least, tells about the surface level findings that this paper has. The mean for the Euro daily foreign exchange returns (against the USD) is 0.707757866, and for CNY is 7.102536252. With this, the central location of the data – or the average data – on FX involving the Euro and the Yuan is located. Describing the spread of the data, the standard deviation for Euro is pegged at 0.04879804; for the Yuan, it is 0.396140195. The variance of the daily return of Euro is 0.002380835, while that of the Yuan is 0.157142022. While the variance results do not constitute an immediate economic interpretation, they make it appear that the variables or components of the population are stable. The correlation returns between Euro and CNY are negative – that is, -0.072435066. It would mean that there is no relationship between the two currencies. And, should the result is stretched to account for their relationship, the two may be said to have negative relationship. Now, this kind of result seems to suggest that volatility spillover effects may not be strongly present between these currencies (Treepongkaruno & Gray, 2009). Discussion and conclusion The mean values of EUR and CNY lend themselves very handily in determining whether these currencies appreciated or depreciated against the USD during the occasion of the 2007-2009 global financial crises. They can just be compared to FX rates on any period before or after the crisis. The standard deviation results point to the fact that the Euro is less volatile than the Yuan – that is, the Euro had stayed closer to its mean value. Very significantly, an indication has been found that correlation between the daily FX returns of the two currencies does suggest that volatility spillover effects may not be that significant between the Euro and the Yuan. Following the model developed by Fleming, Kirby & Ostdek (1998), the connection between the two currencies must have been effected by common information between them and the spillover from one currency to the other as driven specifically by hedging demands. Now, it would mean that stronger links between currencies are the product of the higher levels of information and higher cross-hedging demands (Treepongkaruno & Gray, 2009). Interestingly, though, between Yuan and the Euro (wit USD as the base currency), there seems to be less connection. For one, during the 2007-2009 global financial meltdown, it was the Western counties that were most hardly hit. The global financial crisis, in particular, put the Euro under stress (Jones, 2009). In contrast, China and some countries in Asia stood their ground and were even instrumental in rehabilitating the global economy. Tiding the global financial mess over, China was known to have powerful tools. First, they have largest cash reserve in the world to the tune of 2.447 billion dollars. Second, their Yuan is undervalued by nearly forty per cent (40%) (Nie, 2008). Now, the undervaluing of Yuan and/or the narrowing of the fluctuations of the RMB exchange rate are said to be intentional on the part of China since these constituted their attempt to stabilize market sentiments and stimulate economic activity amidst the crisis. The result was, during the worst of the crises, FX of other currencies to the USD depreciated significantly or by large margins while the Yuan stabilized. In effect, against these depreciating currencies, the value of the Chinese currency rose. To date, however, there have been indications that China will allow more flexibility to its Yuan exchange rate. Xinhua (2010) opines that this move is tantamount to putting an end to the crisis-mode policy that the Chinese government has adopted to protect itself and its people from the brunt of the most recent global financial crisis. References: Antell, J. 2004. Volatility linkages in the Finnish stock, bond and money markets. EFMA 2004 Basel Meetings Paper. Available at: http://ssrn.com/abstract=495082 [Accessed 2 October, 2010]. Bank of International Settlement (BIS). 2010. Triennial Central Bank survey of foreign exchange and derivative market activity in 2007. Available at: http://www.bis.org/publ/rpfxf07t.pdf [Accessed 3 October, 2010]. Bollerslev, T. 1986. Generalized autoregressive conditional heteroscedasticity. Journal of Econometrics, 31, pp. 307-327. Cai, Y. & Treepongkaruna, S. (n.d.). The dynamic relations among volume, return and volatility in the FX spot market. Available at: http://www.fma.org/Xiamen/FMA-Asia_Yiyong_Sirimon.pdf [Accessed 2 October 2010]. Cheong, C.W., Nor, A.H. & Isa, Z., 2007. Long persistence volatility and links between national stock market indices. International Research Journal of Finance and Economics, 7, pp. 175-195. Christensen, C., 2004. Decomposing European bond and equity volatility. International Journal of Finance and Economics, 15 (2), pp. 105-122. Dacorogna, M.M. et al., 1993. A geographical model for the daily and weekly seasonal volatility in the foreign exchange market. Journal of International Money and Finance, 12, pp. 413-438. Ding, Z., Engle, R.F., & Granger, C.W.J., 1993. Long memory properties of stock market returns and a new model. Journal of Empirical Finance, 1, pp. 83-106. Engle, R.F., 1982. Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflation. Econometrica, 50 (4), pp. 987-1008. Engle, R.F. & Kroner, K.F., 1995. Multivariate simultaneous generalized ARCH. Econometric Theory, 11 (1), pp. 122-150. Engle, R.F., Lilien, D.M. & Robins, R.P., 1987. Estimating time varying risk premia in the term structure: the ARCH-M model. Econometrica, 55, pp. 391-407. Federal Reserve Statistical Release (FRSR), 2009 (June 1). Foreign exchange rates (monthly). Available at: http://www.federalreserve.gov/releases/g5 [Accessed 1 October 2010]. Fleischer, P. 2003. Volatility and information linkages across markets and countries. Australian Journal of Management, 28 (3), pp. 251-272. Fleming, J., Kirby, C. & Ostdek, B., 1998. Information and volatility links in stock, bond, and money markets. Journal of Financial Economics, 49, pp. 111-137. Fung, H.G. & Patterson, G.A., 1999. The dynamic relationship of volatility, volume and market depth in currency futures markets. Journal of International Financial Markets, 9 (1), pp. 33-59. Glosten, L.R., Jagannathan, R. & Runkle, D., 1993. On the relation between the expected value and the volatility of the nominal excess return on stocks. Journal of Finance, 48, pp. 1779-1801. Goodhart, C., Ito, T. & Payne, R., 1996. One day in June 1993: a study of the working of Reuters 2000-2 electronic foreign exchange trading system. In J. Frankel, G. Galli & A. Giovannini (Eds.), The microstructure of foreign exchange markets, pp. 107-179. Chicago: University of Chicago Press. Hall, B., 1999. Notes on Generalized Method of Moments estimation. Available at: http://www.nuffield.ox.ac.uk/users/hall/gmmnotes.pdf [Accessed 5 October, 2010]. Jones, E., 2009. The Euro and the financial crisis. Survival, 51 (2), pp. 41-54. Kocagil, A.E. & Schachmurove, Y., 1998. Volume-return dynamics in future markets. Journal of Futures Markets, 18 (4), pp. 399-426. Lien, K., 2010. Forex trading volume officially hits $4 trillion. BK Forex Advisor. Available at: http://www.bkforexadvisors.com/kathy-lien/forex-trading-volume-officially-hits-4-trillion/ [Accessed 4 October 2010]. Nelson, D.B., 1991. Conditional heteroskedasticity in asset returns: a new approach. Econometrica, 59, pp. 347-370. Nguyen, D. & Daigler, R., 2006. A return-volume-volatility analysis of future contracts. Review of Future Markets, 15 (3), pp. 265-293. Nie, P., 2008. Exports to US slow down due to subprime crisis, appreciating Yuan. Available at: http://www.chinadaily.com.cn/bizchina/2008-09/23/content_7052540.htm [Accessed 5 October, 2010]. OANDA, 2010. Average exchange rates. Available at: (see OANDA, 2010. Average exchange rates. Available at: http://www.oanda.com/currency/average [Accessed 5 October 2010]. Ogum, G., 2010. Equity volatility transmission and contagion between the US and the emerging stock markets: the role of the US subprime crisis. Paper presented at the Southwestern Finance Association, 49th Annual Meeting, Dallas, Texas, 2-6 March 2010. Available at: http://southwesternfinance.org/conf-2010/G5-2.pdf [Accessed 2 October, 2010]. Olowe, R.A., 2009. Stock return, volatility and the global financial crisis in an emerging market: the Nigerian case. International Review of Business Research Papers, 5 (4), pp. 426-447. Schwert, W., 1989. Stock volatility and crash of ’87. Review of Financial Studies, 3, pp. 77-102. Taylor, S., 1986. Modeling financial time series. London: John Wiley & Sons. Tauchen, G.E. & Pitts, M., 1983. The price variability-volume relationship on speculative market. Econometrica, 51, pp. 485-505. Treepongkaruna, S. & Gray, S., 2009. Information and volatility links in the foreign exchange market. Accounting and Finance, 49, pp. 385-405. Xinhua, 2010. Yuan move signals end to crisis-mode policy. Available at: http://www.china.org.cn/business/2010-06/22/content_20315283.htm [Accessed 5 October, 2010]. Zakoian, J.M. 1994. Threshold heteroskedastic models. Journal of Economic Dynamics and Control, 18, pp. 931-944. Read More