{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

FINM345A09Lecture2Corr

# FINM345A09Lecture2Corr - FinM 345/Stat 390 Stochastic...

This preview shows pages 1–7. Sign up to view the full content.

FinM 345/Stat 390 Stochastic Calculus, Autumn 2009 Floyd B. Hanson , Visiting Professor Email: [email protected] Master of Science in Financial Mathematics Program University of Chicago Lecture 2 (Corrected October 6, 2009) 6:30-9:30 pm, 05 October 2009, Kent 120 in Chicago 7:30-10:30 pm, 05 October 2009 at UBS Stamford 8:30-11:30 am, 06 October 2009 at Spring in Singapore FINM 345/Stat 390 Stochastic Calculus Lecture2–page1 Floyd B. Hanson

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
Continuation of Lecture 1: 1.4. Time-Dependent (NonHomogeneous) Poisson Process: Financial markets are very time-dependent, so modelers need to think critically about constant coefficient models, understanding that in some cases time-dependence of coefficients may be difficult to estimate, but perhaps not much more difficult to analyze. Thus, consider λ = λ ( t ) so the Poisson process P(t) will be nonstationary . Thus, the Poisson parameter differential is d Λ( t ) λ ( t ) dt , while the integral parameter, assuming Λ(0) = 0 as in the constant jump rate case, is Λ( t ) = t 0 λ ( s ) ds. FINM 345/Stat 390 Stochastic Calculus Lecture2–page2 Floyd B. Hanson
Then, the Poisson parameter increment is defined by ΔΛ( t ) Λ( t + Δ t ) - Λ( t ) = t t t λ ( s ) ds. Thus, ΔΛ( t ) λ ( t t only when Δ t 1 , i.e., is small, but if not the integral must be used. The temporal Poisson distributions Prob[ dP ( t ) = k ] = p k [1:3] ( t )) for the three cases Δ P [1:3] ( t ) = [ dP ( t ) , Δ P ( t ) , P ( t )] and parameters ΔΛ [1:3] ( t ) = [ d Λ( t ) , ΔΛ( t ) , Λ( t )] , are the same Φ Δ P i ( t ) ( k ; ΔΛ i ( t )) = e - ΔΛ i ( t ) (ΔΛ i ( t )) k k ! , for i = 1:3 and k = 0, 1, 2, . . . jumps, t 0 and Δ t 0 . ( Hint: In MATLAB, 1: n =[ j ] 1 × n is a row-vector. ) FINM 345/Stat 390 Stochastic Calculus Lecture2–page3 Floyd B. Hanson

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
Note that all three Poisson processes are increment processes, even Δ P 3 ( t ) = P ( t ) = P ( t ) - P (0) , where P (0) 0 . Also, Λ( t ) is continuous as integrals with λ ( t ) > 0 for t > 0 . While the basic statistics for the set of Poisson increment processes are similar to the simple constant rate case, i.e., E[Δ P i ( t )] = ΔΛ i ( t ) = Var[Δ P i ( t )] . However, the exponential distribution of the interjump times are much more complicated, but see Hanson’s (2007), pp. 22-23, and cited background references. Some theory becomes more complicated in the nonstationary case, but changing the clock from t to Λ( t ) , by changing the rate to the constant one, will remove most of the difficulties (see text, p. 22). FINM 345/Stat 390 Stochastic Calculus Lecture2–page4 Floyd B. Hanson
1.5. Martingale Properties of Markov Processes — Expectations Conditioned on the Past: Simple Definition 1.7: A martingale M(t) is a stochastic process that principally satisfies E[ M ( t ) | M ( s ) , 0 s < t ] = M ( s ) , with some technical side conditions in probability space that M(t) is absolutely integrable , i.e., E[ | M ( t ) | ] < on [0,T] for some finite horizon time T < . (Comment: The term Martingale comes from horse racing and abstractly symbolizes a fair game since E [ M ( t ) - M ( s ) | M ( s )] = 0 , 0 s < t, i.e., there being no net gain on the average conditioned on past data. Alternately, E M ( t ) | M ( t )] = 0 , t 0 .) FINM 345/Stat 390 Stochastic Calculus Lecture2–page5 Floyd B. Hanson

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}