# Difference between revisions of "Published Papers"

(New page: {{pdf|ACFM2001.pdf|Multi-Step estimation of Multivariate GARCH models}}, <i>Proceedings of the International ICSC Symposium: Advanced Computing in Financial Markets</i>, June 2001.<br /> E...) |
|||

Line 1: | Line 1: | ||

− | {{pdf|ACFM2001.pdf|Multi-Step estimation of Multivariate GARCH models}}, | + | {{pdf|ACFM2001.pdf|Multi-Step estimation of Multivariate GARCH models}}, ''Proceedings of the |

− | International ICSC Symposium: Advanced Computing in Financial Markets | + | International ICSC Symposium: Advanced Computing in Financial Markets'', June 2001.<br /> |

Estimation of large time varying covariance matrices has proven difficult since the original multivariate volatility models were introduced. In this paper, we develop the empirical properties of a new class of MV-GARCH models capable of estimating large time-varying covariance matrices. We show that the problem of multivariate conditional variance estimation can be reduced to a series of univariate GARCH processes plus an additional conditional correlation estimator. We use the model to estimate a conditional covariance of up to 100 assets using S&P 500 Sector Indies and Dow Jones Industrial Average stocks. This new estimator demonstrates very strong performance especially considering the estimators easy of implementation. | Estimation of large time varying covariance matrices has proven difficult since the original multivariate volatility models were introduced. In this paper, we develop the empirical properties of a new class of MV-GARCH models capable of estimating large time-varying covariance matrices. We show that the problem of multivariate conditional variance estimation can be reduced to a series of univariate GARCH processes plus an additional conditional correlation estimator. We use the model to estimate a conditional covariance of up to 100 assets using S&P 500 Sector Indies and Dow Jones Industrial Average stocks. This new estimator demonstrates very strong performance especially considering the estimators easy of implementation. | ||

− | {{pdf|ordering2003.pdf|An Ordering Experiment}}, | + | {{pdf|ordering2003.pdf|An Ordering Experiment}}, ''Journal of Economic Behavior and Organization'' with A. Norman, et |

al.,February 2003, pp. 249-262.<br /> | al.,February 2003, pp. 249-262.<br /> | ||

Binary comparison operators form the basis of consumer set theory. If humans could only perform binary comparisons, the most efficient procedure a human might employ to make a complete preference ordering of n items would be a n log<sub>2</sub>n algorithm. But, if humans are capable of assigning each item an ordinal utility value, they are capable of implementing a more efficient linear algorithm. In this paper, we consider six incentive systems for ordering three different sets of objects: pens, notebooks, and Hot Wheels. All experimental evidence indicates that humans are capable of implementing a linear algorithm, for small sets.</td> | Binary comparison operators form the basis of consumer set theory. If humans could only perform binary comparisons, the most efficient procedure a human might employ to make a complete preference ordering of n items would be a n log<sub>2</sub>n algorithm. But, if humans are capable of assigning each item an ordinal utility value, they are capable of implementing a more efficient linear algorithm. In this paper, we consider six incentive systems for ordering three different sets of objects: pens, notebooks, and Hot Wheels. All experimental evidence indicates that humans are capable of implementing a linear algorithm, for small sets.</td> |

## Revision as of 23:24, 20 August 2007

Multi-Step estimation of Multivariate GARCH models , *Proceedings of the*
International ICSC Symposium: Advanced Computing in Financial Markets*, June 2001.*
Estimation of large time varying covariance matrices has proven difficult since the original multivariate volatility models were introduced. In this paper, we develop the empirical properties of a new class of MV-GARCH models capable of estimating large time-varying covariance matrices. We show that the problem of multivariate conditional variance estimation can be reduced to a series of univariate GARCH processes plus an additional conditional correlation estimator. We use the model to estimate a conditional covariance of up to 100 assets using S&P 500 Sector Indies and Dow Jones Industrial Average stocks. This new estimator demonstrates very strong performance especially considering the estimators easy of implementation.

An Ordering Experiment , *Journal of Economic Behavior and Organization* with A. Norman, et
al.,February 2003, pp. 249-262.

Binary comparison operators form the basis of consumer set theory. If humans could only perform binary comparisons, the most efficient procedure a human might employ to make a complete preference ordering of n items would be a n log_{2}n algorithm. But, if humans are capable of assigning each item an ordinal utility value, they are capable of implementing a more efficient linear algorithm. In this paper, we consider six incentive systems for ordering three different sets of objects: pens, notebooks, and Hot Wheels. All experimental evidence indicates that humans are capable of implementing a linear algorithm, for small sets.
On the Computational Complexity of Consumer Decision Rules , 2003, with A. Norman, et. al., *Computational Economics*, Vol 23, March 2004, pp 173-192.

A consumer entering a new bookstore can face more than 250,000 alternatives. The efficiency of compensatory and noncompensatory decision rules for finding a preferred item depends on the efficiency of their associated information operators. At best, item-by-item information operators lead to linear computational complexity; set information operators, on the other hand, can lead to constant complexity. We perform an experiment demonstrating that subjects are approximately rational in selecting between sublinear and linear rules. Many markets are organized by attributes that enable consumers to employ a set-selection-by-aspect rule using set information operations. In cyberspace decision rules are encoded as decision aids.