Optimalsellingrulesinaregimeswitchingmodel,-1450.PDFVIP

1450 IEEE TRANSACTIONS ON AUTOMATIC CONTROL, VOL. 50, NO. 9, SEPTEMBER 2005 REFERENCES I. INTRODUCTION [1] B. D. O. Anderson, M. Deistler, L. Farina, and L. Benvenuti, “Nonneg- Consider an investor who holds a share of stock whose price at time ative realization of a linear system with nonnegative impulse response,” t ? @ t A ( IEEE Trans. Circuits Syst. I, Fundam. Theory Appl., vol. 43, no. 2, pp. is . The investor’s goal is to ?nd the “best” selling time to 134–142, Feb. 1996. maximize the discounted expected payoff, i.e., the value function is r( [2] B. D. O. Anderson, “New developments in the theory of positive sys- m?x ( I i x ? e @ ? ( u A? , where u isthe amount to be paid back tems,” in Systems and Control in the Twenty-First Century. Boston, when the investor sells the stock, ? @HA a x , and r isthe discountfactor MA: Birkh?user, 1997. (e.g., due to in?ation). Assume also that the investor is not clairvoyant. [3] L. Benvenuti and L. Farina, “A tutorial on the positive realization ( t problem,” IEEE Trans. Autom. Control, vol. 49, no. 5, pp. 651–664, This is an optimal stopping time problem with the stopping time May 2004. measurable to the “information” available up to time t (characterized [4] , “The design of ?ber-optic ?lters,” J. Lightw. Technol., vol. 19, no. by the -algebra generated by ? @ s A Y H s t ). Clearly, the choice 9, pp. 1366–1375, Sep. 2001.