A New Method for Estimating and Simulating Maximum Entropy Densities
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
The maximum entropy (maxent) approach provides an attractive and appealing method for obtaining a probability distribution in some limited information environments. The maxent approach, however, is not easy to use for practical application since analytic derivation of a maxent density is usually not feasible. This paper proposes a method of estimating and simulating an maxent density. It is a numerically tractable and stable method that is relatively simple in real computation. The proposed method is capable of handling problems for which existing methods are difficult to apply or are subject to occasional failure. Monte Carlo results confirm that the proposed method can be well applied for cases when existing methods are subject to occasional failure.
Abstract
The maximum entropy (maxent) approach provides an attractive and appealing method for obtaining a probability distribution in some limited information environments. The maxent approach, however, is not easy to use for practical application since analytic derivation of a maxent density is usually not feasible. This paper proposes a method of estimating and simulating an maxent density. It is a numerically tractable and stable method that is relatively simple in real computation. The proposed method is capable of handling problems for which existing methods are difficult to apply or are subject to occasional failure. Monte Carlo results confirm that the proposed method can be well applied for cases when existing methods are subject to occasional failure.
참고문헌 (15)
- Andersen, P. K., "Cox’s regression model for counting processes: A large sample study", Annals of Statistics, 10, pp.1100~1120 (1982)
- Maximum entropy autoregressive conditionial heteroskedasticity model, "Bera, A. K.", University of Illinois at Urbana-Champaign, ~ (2004)
- Buchen, P., "The maximum entropy distribution of an asset inferred from option prices", Journal of Financial and Quantitative Analysis, 31, pp.143~159 (1996)
- Golan, A., "Maximum Entropy Econometrics: Robust Estimation with Limited Data" (1996), John Wiley and Sons
- Jaynes, E. T., "Information theory and statistical mechanics", Physics Review, 106, pp.620~630 (1957)
- Kim, J. Y., "Limited information likelihood and Bayesian analysis", Journal of Econometrics, 107, pp.175~193 (2002)
- Lagarias, J. C., "Convergence properties of the Nelder-Mead simplex algorithm in low dimensions", SIAM Journal on Optimization, 9, pp.112~147 (1998)
- Ormoneit, D., "An efficient algorithm to compute maximum entropy densities", Econometrics Reviews, 18, pp.127~140 (1999)
- Pollard, D., "Asymptotics for least absolute deviation regression estimators", Econometric Theory, 7, pp.186~199 (1991)
- Rockinger, M., "Entropy densities with an application to autoregressive conditional skewness and kurtosis", Journal of Econometrics, 106, pp.119~142 (2002)
- Stutzer, M., "A simple nonparametric approach to derivative security valuation", Journal of Finance, 51, pp.1633~1652 (1996)
- Van der Vaart, A.W., "Asymptotic Statistics" (1998), Cambridge University Press
- Wu, X., "Calculation of maximum entropy densities with application to income distribution", Journal of Econometrics, 115, pp.347~354 (2003)
- Zellner, A., "The Bayesian method of moments (BMOM): Theory and applications, in Advances in Econometrics (Applying Maximum Entropy to Econometric Problems), Vol. 12", pp.85~105 (1997), Emerald Group Publishing Limited
- Zellner, A., "Calculation of maximum entropy distribution and approximation of marginal posterior distributions", Journal of Econometrics, 37, pp.195~209 (1988)