In this paper we suggest using the FACTOR-ARCH model as a parsimonious structure for the conditio... more In this paper we suggest using the FACTOR-ARCH model as a parsimonious structure for the conditional covariance matrix of asset excess returns. This structure allows us to study the dynamic relationship between asset risk premia and volatilities in a multivariate system. One and two FACTOR-ARCH models are succussfully applied to pricing of Treasury bills. The results show stability over time, pass a variety of diagnostic tests, and compare favorably with previous empirical findings. 1.
What determines the volatility of asset prices? Most economists have traditionally argued that fu... more What determines the volatility of asset prices? Most economists have traditionally argued that fundamental factors determine volatility. Others assert, however, that asset prices are driven by "animal spirits" and other random forces which induce excess volatility.1 In this article ...
This paper studies asymmetric responses of covariances to return shocks. Asymmetric/leverage effe... more This paper studies asymmetric responses of covariances to return shocks. Asymmetric/leverage effects have been found in variances but few studies have examined such effects in covariances, even though they might have important implications for portfolio management and hedging. We propose a robust conditional moment test to detect the presence of such asymmetric effects in the covariances. We also introduce a general dynamic covariance matrix model, which nests many of the existing multivariate GARCH models and addresses covariance asymmetries. To illustrate the tests and the new model, we apply them to weekly returns from a large firm and a small firm portfolio. Our results suggest that bad news about large firms can cause volatility in both small firm returns and large firm returns. The conditional covariance between large firm returns and small firm returns will also increase following bad news about large firms. In contrast, news about small firms has a minimal effect on variance...
Existing multivariate models allowing the covariance matrix to be time varying generally impose s... more Existing multivariate models allowing the covariance matrix to be time varying generally impose strong restrictions on how past shocks can affect the covariance matrix. Yet these restrictions are seldom compared and tested. Furthermore, few studies have examined the existence of asymmetric/leverage effects in covariance, even though there are good reasons to believe they exist and have important implications for portfolio management, derivatives pricing and hedging. In this paper, we introduce a set of robust conditional moment tests to detect misspecification in the covariance matrix, and introduce a general model which nests the four most popular multivariate GARCH models and their natural "asymmetric" extensions. The new model is applied to study the dynamic relation between large and small firm returns. Our results suggest that bad news about large firms can cause volatility in both small and large firm returns, while news about small firms, whether good or bad, has mi...
... restricted); Francisco Climent Diranzo & Robert Meneu Gaya, . &qu... more ... restricted); Francisco Climent Diranzo & Robert Meneu Gaya, . "Relaciones de equilibrio entre demografía y crecimiento económico en España," Studies on the Spanish Economy 163, FEDEA. [Downloadable!]; Raj Aggarwal & Brian M. Lucey & Cal Muckley, 2004. ...
Page 1. Inferring Future Volatility from the Information in Implied Volatility in Eurodollar Opti... more Page 1. Inferring Future Volatility from the Information in Implied Volatility in Eurodollar Options: A New Approach Kaushik I. Amin Lehman Brothers Victor K. Ng Goldman Sachs & Co. We study the information content of implied ...
Existing time-varying covariance models usually impose strong restrictions on how past shocks aff... more Existing time-varying covariance models usually impose strong restrictions on how past shocks affect the fore-casted covariance matrix. In this article we compare the restrictions imposed by the four most popular mul-tivariate GARCH models, and introduce a set of robust ...
ABSTRACT We use an extension of the equilibrium framework of Rubinstein (1976) and Brennan (1979)... more ABSTRACT We use an extension of the equilibrium framework of Rubinstein (1976) and Brennan (1979) to derive an option valuation formula when the stock return volatility is both stochastic and systematic. Our formula incorporates a stochastic volatility process as well as a ...
What determines the volatility of asset prices? Most economists have traditionally argued that fu... more What determines the volatility of asset prices? Most economists have traditionally argued that fundamental factors determine volatility. Others assert, however, that asset prices are driven by "animal spirits" and other random forces which induce excess volatility.1 In this article ...
... Victor K. NG Research Department, International Monetary Fund, Washington, DC 20431 ... In ad... more ... Victor K. NG Research Department, International Monetary Fund, Washington, DC 20431 ... In addition, Clemons and Weber (1989) reported that nearly 50% of trans-actions on the LSE by market value occur within the touch, but a 1992 study by the LSE's Quality of Markets ...
ABSTRACT We explore the time series properties of stock returns on the London Stock Exchange arou... more ABSTRACT We explore the time series properties of stock returns on the London Stock Exchange around the 1986 market restructuring (Big Bang) and the 1987 stock-market crash using a modified generalized autoregressive conditional heteroscedasticity model. Using this general dynamic model, which allows (a) intradaily returns to have different impacts and persistence on stock-return volatility, (b) return effects on volatility to be asymmetric, and (c) intradaily returns to follow conditional distributions with different fourth moments, we uncover important changes in return dynamics and conditional fourth moments following Big Bang and the 1987 crash not reported before.
In this paper we suggest using the FACTOR-ARCH model as a parsimonious structure for the conditio... more In this paper we suggest using the FACTOR-ARCH model as a parsimonious structure for the conditional covariance matrix of asset excess returns. This structure allows us to study the dynamic relationship between asset risk premia and volatilities in a multivariate system. One and two FACTOR-ARCH models are succussfully applied to pricing of Treasury bills. The results show stability over time, pass a variety of diagnostic tests, and compare favorably with previous empirical findings. 1.
What determines the volatility of asset prices? Most economists have traditionally argued that fu... more What determines the volatility of asset prices? Most economists have traditionally argued that fundamental factors determine volatility. Others assert, however, that asset prices are driven by "animal spirits" and other random forces which induce excess volatility.1 In this article ...
This paper studies asymmetric responses of covariances to return shocks. Asymmetric/leverage effe... more This paper studies asymmetric responses of covariances to return shocks. Asymmetric/leverage effects have been found in variances but few studies have examined such effects in covariances, even though they might have important implications for portfolio management and hedging. We propose a robust conditional moment test to detect the presence of such asymmetric effects in the covariances. We also introduce a general dynamic covariance matrix model, which nests many of the existing multivariate GARCH models and addresses covariance asymmetries. To illustrate the tests and the new model, we apply them to weekly returns from a large firm and a small firm portfolio. Our results suggest that bad news about large firms can cause volatility in both small firm returns and large firm returns. The conditional covariance between large firm returns and small firm returns will also increase following bad news about large firms. In contrast, news about small firms has a minimal effect on variance...
Existing multivariate models allowing the covariance matrix to be time varying generally impose s... more Existing multivariate models allowing the covariance matrix to be time varying generally impose strong restrictions on how past shocks can affect the covariance matrix. Yet these restrictions are seldom compared and tested. Furthermore, few studies have examined the existence of asymmetric/leverage effects in covariance, even though there are good reasons to believe they exist and have important implications for portfolio management, derivatives pricing and hedging. In this paper, we introduce a set of robust conditional moment tests to detect misspecification in the covariance matrix, and introduce a general model which nests the four most popular multivariate GARCH models and their natural "asymmetric" extensions. The new model is applied to study the dynamic relation between large and small firm returns. Our results suggest that bad news about large firms can cause volatility in both small and large firm returns, while news about small firms, whether good or bad, has mi...
... restricted); Francisco Climent Diranzo & Robert Meneu Gaya, . &qu... more ... restricted); Francisco Climent Diranzo & Robert Meneu Gaya, . "Relaciones de equilibrio entre demografía y crecimiento económico en España," Studies on the Spanish Economy 163, FEDEA. [Downloadable!]; Raj Aggarwal & Brian M. Lucey & Cal Muckley, 2004. ...
Page 1. Inferring Future Volatility from the Information in Implied Volatility in Eurodollar Opti... more Page 1. Inferring Future Volatility from the Information in Implied Volatility in Eurodollar Options: A New Approach Kaushik I. Amin Lehman Brothers Victor K. Ng Goldman Sachs & Co. We study the information content of implied ...
Existing time-varying covariance models usually impose strong restrictions on how past shocks aff... more Existing time-varying covariance models usually impose strong restrictions on how past shocks affect the fore-casted covariance matrix. In this article we compare the restrictions imposed by the four most popular mul-tivariate GARCH models, and introduce a set of robust ...
ABSTRACT We use an extension of the equilibrium framework of Rubinstein (1976) and Brennan (1979)... more ABSTRACT We use an extension of the equilibrium framework of Rubinstein (1976) and Brennan (1979) to derive an option valuation formula when the stock return volatility is both stochastic and systematic. Our formula incorporates a stochastic volatility process as well as a ...
What determines the volatility of asset prices? Most economists have traditionally argued that fu... more What determines the volatility of asset prices? Most economists have traditionally argued that fundamental factors determine volatility. Others assert, however, that asset prices are driven by "animal spirits" and other random forces which induce excess volatility.1 In this article ...
... Victor K. NG Research Department, International Monetary Fund, Washington, DC 20431 ... In ad... more ... Victor K. NG Research Department, International Monetary Fund, Washington, DC 20431 ... In addition, Clemons and Weber (1989) reported that nearly 50% of trans-actions on the LSE by market value occur within the touch, but a 1992 study by the LSE's Quality of Markets ...
ABSTRACT We explore the time series properties of stock returns on the London Stock Exchange arou... more ABSTRACT We explore the time series properties of stock returns on the London Stock Exchange around the 1986 market restructuring (Big Bang) and the 1987 stock-market crash using a modified generalized autoregressive conditional heteroscedasticity model. Using this general dynamic model, which allows (a) intradaily returns to have different impacts and persistence on stock-return volatility, (b) return effects on volatility to be asymmetric, and (c) intradaily returns to follow conditional distributions with different fourth moments, we uncover important changes in return dynamics and conditional fourth moments following Big Bang and the 1987 crash not reported before.
Uploads
Papers by Victor Ng