problemset3

# problemset3 - Solutions to Problem Set 3 Chris H Rycroft 1...

This preview shows pages 1–4. Sign up to view the full content.

Solutions to Problem Set 3 Chris H. Rycroft October 30, 2006 1 Inelastic diffusion 1.1 The PDF of P N ( x ) The PDF of the random variable a n Δ X n is p ( xa n ) a n . If Δ X n has characteristic function ˆ p ( k ) then a n Δ X n has characteristic function e ikx p ( xa n ) dx = e ikya n p ( y ) dy a n −∞ −∞ = p ˆ( ka n ) . By the convolution theorem, we know that the characteristic function of X N can be written as P ˆ N ( k ) = N p ˆ( ka n ) n =1 and hence the PDF is given by e ikx dx P N ( k ) = N 2 π n =1 −∞ p ˆ( ka n ) N = 2 π n =1 −∞ −∞ 1.2 The cumulants of X N e ikx dx e ikya n p ( y ) dy. The cumulant generating function of the n th step is given by ψ n ( k ) = log ˆ p ( ka n ) and therefore the l th cumulant is given by i l d l ψ n dk l = a ln i l d l dk l ψ n = a ln c l . k =0 k =0 If random variables are added, then their cumulants add. We therefore know that the cumulants of X N are given by N C N,l = a l (1 a lN ) ln c l = 1 a l a c l . n =1 1

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

View Full Document
M. Z. Bazant 18.366 Random Walks and Diffusion Problem Set 3 Solutions 2 1.3 Limits as N → ∞ By taking N → ∞ in the above expression, we see that the cumulants of X are given by l a C l = c l . 1 a l If a = 1 where � > 0 is small, then we see that C 2 m a 2 m 1 a 2 m c 2 m = C 2 2 m a 2 c 2 m 1 a m (2 2 ) m c 2 m = 1 (1 ) 2 m c 2 m 2 m m 1 2 m c 2 m = c 2 m� 2 m (2 m 1) 2 + . . . m 2 2 = 2 m m 1 1 �m + . . . c 2 m 2 m 2 m (2 2 m 1) 2 + . . . c m = O ( m 1 ) . 1.4 A “Central Limit Theorem” The PDF of X can be written as P ( x ) = dk e ikx + ψ ( k ) 2 π −∞ dk C 2 k 2 iC 3 k 3 C 4 k 4 = 2 π exp ikx + 0 2 6 + 24 . −∞ If x = ζC 2 1 / 2 we see that 1 / 2 1 / 2 dl l 2 C 3 l 3 C 4 l 4 φ ( ζ, � ) = C 2 P ( ζC 2 ) = 2 π exp ilζ 2 i 6 C 1 / 2 + 24 C 2 2 . −∞ 2 From the previous result, we know that as 0, the terms involving the higher cumulants tend to zero. Therefore we have φ ( ζ, � ) φ ( ζ ) = dl e ilζ l 2 / 2 2 π −∞ e ζ 2 / 2 = . 2 π 2 Breakdown of the CLT for decaying walks 2.1 The PDF of X for a = 0 . 99 Appendix A shows a simple C++ code to simulate 10 5 walkers from the decaying PDF. Figure 1 shows the PDF of x component of this distribution, and figure 2 shows a comparison to the
M. Z. Bazant 18.366 Random Walks and Diffusion Problem Set 3 Solutions 3 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 -30 -20 -10 0 10 20 30 P ( x ) x Figure 1: Simulated PDF of X for a = 0 . 99 based on 10 9 trials. formulae calculated in question 1. We see that the two curves match to a very high degree of accuracy, particularly in the central region. We know however that this match will not continue forever: since X is bounded, its PDF will be uniformly zero outside of some region. However a Gaussian features non-zero probabilities everywhere.

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 ]}

### What students are saying

• As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

Kiran Temple University Fox School of Business ‘17, Course Hero Intern

• I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

Dana University of Pennsylvania ‘17, Course Hero Intern

• The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

Jill Tulane University ‘16, Course Hero Intern