hw6_soln

hw6_soln - Probability and Stochastic Processes: A Friendly...

Info iconThis preview shows page 1. Sign up to view the full content.

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

Unformatted text preview: Probability and Stochastic Processes: A Friendly Introduction for Electrical and Computer Engineers Roy D. Yates and David J. Goodman Problem Solutions : Yates and Goodman,3.5.3 3.6.3 3.6.7 3.7.2 and 3.7.3 n1 1 1n p∑ © ¥ ¦ k p 1 p n1 p ¤ n ¤ 1 ¤ ¨ ¥ ¤ ¡ § ¥ k1 nk ¨ ∑ PN K ¥ ¡ n ¤ PN n , the marginal PMF of N satisfies 1 © 12 ¢¢¢ £££¡ Problem 3.5.3 First we observe that for n ¦ Thus, the event B has probability  ¥ p ¤ ¨ ¤ 1 1 p ¨  p 9p 1 2 ¥  ¥ ¨ ¤ ¥ ¤   n 10 1  ££ n ¨ ∑ PN p ¥ ∞ PB 9 ¦ From Theorem 3.11, PN K n k PB "! § n n 10 11 otherwise ;k ¢¢¢ £££¡ ¡ $ n 10 p 1 ©¥ p ¡ ¢¢¢ %£££¡ ¨ ¥ ¤ ¡ § ¤ 1 0 ¡ 0 nk B otherwise # PN K B n k n ¤ ¤ The conditional PMF PN B n b could be found directly from PN n using Theorem 2.19. However, we can also find it just by summing the conditional joint PMF. n 10 p ©¥ p n 10 11 otherwise ¨ ¤ 1 0 ¡ ¥ ¡ ¤ §  ¥ k1 nk ¢¢¢ £££¡ ¥& ∑ PN K B ¥ n ¤ PN B n ¦ ¤  From the conditional PMF PN B n , we can calculate directly the conditional moments of N given B. Instead, however, we observe that given B, N N 9 has a geometric PMF with mean 1 p. That is, for n 1 2 , $ ¥ ' ¨  ¢¢¢ £££¡ ¤ PN B n 9  9B ¥ &   n ¤ PN 1 p ¨  ¥ ¤ n1 p 9 and we can calculate the conditional expectations 1 ¨ Var N B 1p 9 p p2 $ £¥ &' ' & 9B 9 ¤ &' & ' Var N ENB  9B &  Var N B EN $  0' ENB   )( N n ¥ Hence, given B, N B © ¡ PN &  Note that further along in the problem we will need E N 2 B which we now calculate. & ¤  1 &    1 ENB 81  Var N B 2 17 2 p p ¥ 2 & E N2 B 2 & ¤ For the conditional moments of K , we work directly with the conditional PMF PN K B n k . § 1 10 p n pn n n 10 ∑k k1 ¦ ©¥ ¦ ¨ ¤  &  ¦ ¦ ¤ 1 2, $ £¥ nn ∑ n 10 k 1 ∞ 10 p ¤ ¥ ¨ ¤  2  & n1 p n1 © 1 p 1 EN 2 1 1 2p 1B  n  ∑  & ¦ ∞ EKB 5   1k pn n ¨ Since ∑n k 1 ©¥ ∑ ∑k  ¦ We now can calculate the conditional expectation of the sum. 1 2p   ¤ 10 p 9 $ 1p ¤ EKB 5  4¥  & $ ENB 3 2p   3 & KB 14  &   EN  The conditional second moment of K is p 1 EN 6  n 10 ¤ © ¥ p 1 2N  ¨ ¤¥ 1 ¥ 1 ¤ ¤  ¥ ¤  ¤  1 2n 6 ¥  ¤ ¦ & n 10 1 6, we obtain $ £¥ n k1  & ¦ ¦ ∑ n 10 ∑ k2 1B  5¥ & ∞ 1 2n 10 p n pn n ¦ ¨ nn 1 ¦ ©¥ 2 ∞ ∑ n 10 k 1 1k E K2 B pn n ¤ Using the identity ∑n k 1 ¨ n ∑ ∑ k2 ©¥ ∞ E K2 B ¦ Applying the values of E N B and E N 2 B found above, we find that 1 6 2 3 p2  ENB 2 37 6p  &  E N2 B 3 2 3 31  & & E K2 B &   & Thus, we can calculate the conditional variance of K .  ¥ 2 & 10 p 5 12 p2 ¤ pn n 2 ¨ EKB 7 6p  ¨ E K2 B Var K B 6 2 3  & &  To find the conditional correlation of N and K , n 10 ¨ ∑ n 1 p ¥ n1 p∑k k1 ¦ ¦ & n 10 k 1 ∞  ¦ ¦ 1 2, $ £¥ 2 p 1 EN N 2  n 10 ¤ © ¥ p  ¨ 1 1B &  5¥ ¤¥  2 1 1 p2  n 10 nn 9 p  ∑ ¤  ¤ ∞ ¦ E NK B ¤ nn ¨ &  1k ©¥ Since ∑n k 1 ∑ nk ¤ ∑ n © ∞ E NK B 45 ¡ n ¥ ∞ EKB ¦ Problem 3.6.3 (a) First we observe that A takes on the values SA 1 1 while B takes on values from SB 0 1 . To construct a table describing PA B a b we build a table for all possible values of pairs A B . The general form of the entries is 8 ¨ 7 6 ¡ ¥ ¤ ¡ § ¥ ¥ b1 PB A 1 1 PA 1 PB A 1 1 PA 1 ¨ ¥ ¤ ¥ ¨ 2& ¤ ¤ ¥& ¤ ¥  ¤ ¨ ¥ ¤ ¥  ¤ b0 PB A 0 1 PA 1 PB A 0 1 PA 1 ¤ ¨ 2& ¥& ¤ ¤ ¥ ¤ ¡  8 ¡ 6 § ¨  ¡ ¤ PA B a b a 1 a1 Now we fill in the entries using the conditional PMFs PB A b a and the marginal PMF PA a . This yields ¥&  b1 23 13 12 23 $ ¥ $ ¥ ¤ ¥ ¤ $ ¤ ¥ $ $ ¥ ¤ ¥ ¤ $ b0 13 13 12 23 ¥ $ ¥ ¤ ¥ ¡ ¤ $ ¥ ¤ PA B a b a 1 a1 ¨ ¤ § which simplifies to $ ¥ $ $ ¤ ¡ $ § ¨ 1, then the conditional expectation of B is ¤ b1 PB A 1 1 ¥&    b0 ¤ ∑ bPB A 1 12 ¥& 1 EBA $ (b) If A b 0b 1 19 29 13 13 PA B a b a 1 a1 &  ¦ ¤ (c) Before finding the conditional PMF PA B a 1 , we first sum the columns of the joint PMF table to find 0 1 49 b 59 b $  ¥ $ ¤ 1 is 25 a 35 a ¥ $ 1 1 $ ¤ PA B a 1 PB 1 ¡ § ¤ ¥ ¤ ¤ PA B a 1 ¨ The conditional PMF of A given B ¥& PB b ¥&  (d) Now that we have the conditional PMF PA B a 1 , calculating conditional expectations is easy. ¤ ¥& ¨  ¤ ¥& $  2 $ 1 35 35 15 1 ¤ $ 25  ¥ © 9¦ © 9¦ § a2 PA B a 1 ¤ 11 ∑ 125 $  §  & a aPA B a 1 11 ¥ 1 a $   E A2 B ∑ 1 ¥& EAB & The conditional variance is then ¨ 1 ¤ ¥ 2 &  ¤ 3 EAB 15 $ ¨ 1 ¥ E A2 B 2 & 1 24 25 $  Var A B &  (e) To calculate the covariance, we need ¤ $ ¥ ¨ ¤ 13 §  4¥ ¤ ¤ 11 1 3 $ ¥  4¥ ¤ ¤ 10 13 19 ¥ $ ¥  ¥ ¡ ¤ $ ¤ ¥ ¥ § 11 29 $ ¤ ab 59 ¥ ¤ $ ¤ ¦ ¨ A4¥   ¦ ¤ § © @¦ ¤ 10 19 ¤ 1 1b 0 $ © @¦ 1 159 ¥ 049 $ b ∑ ∑ abPA B $ ¥  a 123  4¥ ∑ bPB b0 E AB ¤ 1 EB 113 $ aPA a 11 ¥ a $ ∑ EA  ¨ The covariance is just 2 27 B ¥ ¤ $ ¨  $  ¨ 1   ¡  Problem 3.6.7 The key to solving this problem is to find the joint PMF of M and N . Note that N the joint event M m N n has probability M. For n 8 © ¥ p nm1 p n22 p © ¥ ¨ p p1 ¨ ¨  ¡  ¤ ¤ 1 m1 ¥ p dv ©   ££ 1 d v dd ¤ P dd  ££ n © © mN © 6 PM m1 calls  n © ¡ m1 calls A complete expression for the joint PMF of M and N is 1; n ¨ n m m 12 otherwise ¡ ¢¢¢ D£££¡ ¡ ©¥ ¤ ¨ ¥ ¤ ¡ § , the marginal PMF of N satisfies n 11 ¨ ¤ p ¥ n22 ¤ © ¥ ¨ ¤ © ¥ m1 p ¨ 1 p ¥ ∑ n22 © n1 ¢¢¢ £££¡ ¡ ¤ PN n ¦ , the marginal PMF of M satisfies ¥ ¨ m1 © ¥ p m1  p 1 p ¨ 1 p ¥ ¤ ¤ ¦  ¨ 1 E p n22 m  ££A ¥ nm1 2 p © ¤ 1 ¤ ∑ ¨ ∞ ¥ ¡ PM m © 12 ¢¢¢ £££¡ Similarly, for m 1m  n 2 p2 ¡ 23 p p ¤ The complete expressions for the marginal PMF’s are ©¥ ¨ 4 n 2 p2 n 23 otherwise p ¡ © ¤ ¥ 11 m 12 otherwise ¢¢¢ £££¡ m 1p ¤ ¥ ¨ ¨ ¥ ¥ n 0 p ¡ ¤ ¤ ¤ PN n 1 0 ¢¢¢ £££¡ PM m p  For n 1 0 2 ¢¢¢ £££¡ PM N m n C ¤ 13 59 $ 19 ¥ EAEB ¨ E AB $ Cov A B m, Not surprisingly, if we view each voice call as a successful Bernoulli trial, M has a geometric PMF since it is the number of trials up to and including the first success. Also, N has a Pascal PMF since it is the number of trials required to see 2 successes. The conditional PMF’s are now easy to find. © ©¥ ¨ ¤ ¥ ¤ ¡ § ¤ ¥ ¥ ¤ &  The interpretation of the conditional PMF of N given M is that given M m, N has a geometric PMF with mean 1 p. The conditional PMF of M given N is m N where N  2 ' n m 1m otherwise ' n m 1p  p ¡ 1 0  PM N m n PM m ¢¢¢ £££¡ PN M n m $ m1 n otherwise ¥ 1 ¡ ¢¢¢ D£££¡ ¨ ¤ 1n 0 ¨ ¥ $ ¤ PM N m n PN n ¡ § ¤ ¥ ¤ PM N m n 1 ¥&  Given that call N n was the second voice call, the first voice call is equally likely to occur in any of the previous n 1 calls. ¨ Problem 3.7.2 Using the following probability model $ ¥ $ ¤ PY k ¥ ¤ PX k 34 k 0 1 4 k 20 0 otherwise We can calculate the requested moments.  2 75 10  $  2E X 5 ¨ 1 4 20 ¥ ¥   ¨    ¤ EX 2 5 5 ¤  EX   $   3 $    Y 34 0 $  EX 1 4 20  Var X 340 EX  Since X and Y are independent, Theorem 3.14 yields ¤ 150 ¤ PX x PY y and ¤ ¤ 20 20 220 20 PX 20 PY 20  ¥ 1012 G ¥ ¤ 2 75 ¥ ¢ ¥ ¤ ¤ ¡  ¡ § § x 0 20 y 0 20 ¤ ¥ XY 2XY PX Y x y ¥ ¥  ∑∑ 2 Var X ¥   E XY 2XY  F Since X and Y are independent, PX Y x y Var Y  Var X  Y  Var X § ¦ § ¦ Problem 3.7.3 (a) Normally, checking independence requires the marginal PMFs. However, in this problem, the zeroes in the table of the joint PMF PX Y x y allows us to verify very quickly that X and Y are dependent. In particular, PX 1 1 4 and PY 1 14 48 but ¤ 1 PY 1 ¥ ¨ ¤ H § ¥ 5 PX ¥ ¤ ¥ $ 0 ¥ ¡ 11 $ ¥ ¤ ¡ ¤ ¨ ¨ ¤ PX Y § ¤ ¤ (b) To fill in the tree diagram, we need the marginal PMF PX x and the conditional PMFs PY X y x . By summing the rows on the table for the joint PMF, we obtain ¥ x 4 2 4 ¤ $ $ $ $ ¤ ¤ ¤ ¥& 13 y 101 0 otherwise $ %¥ ¡ ¤ ¡ ¨ $ $ ¥ $ $ $ PX Y x y PX x to write ¥ ¨ $ § §  ¨ $ ¥& ¥& ¨  ¡ $ $ $ ¨ I& ¤ ¤  PY X y 1 PY X y 0 ¡ ¥ 12 y 01 0 otherwise 1 ¡ 34 y 1 14 y 0 0 otherwise ¤ PX 1 1 1 ¥ 1y 0y 1 3 16 1 16 0 16 16 16 0 18 18 Now we use the conditional PMF definition PY X y x PY X y  y ¥& PX Y x y x 1 x0 x1  Now we can us these probabilities to label the tree. The generic solution and the specific solution with the exact values are 13 Y0 c aa Y1 Q Y0 Q Y1 Q ` bb ` a Q a a a c 12 c b 12 1 Q ` c b b b Q P c VV V V V P V R SQ Q Q c Q Q 6 c` Y1 X1 Y 13 13 14 Y0 X0 12 1 Y0 14 R SQ ` `` U Y ` RT X ` a ` Q UY TX aa a U Y T aX a UY TX bb b bb UY TX PY X 1 1 P PY X 0 1 1 `` R SQ PP U PRT b Q P P UT V X1 Q T VV U PX 1 Y1 cP PY X 1 0 Y0 R SQ PY X 0 0 1 X c X0 Y 14 Y Q U RY T X W WW WW U W R Y R T X PX 0 10 R SQ PY X Y0 PP 1 W PY X 0 W 1 34 1 W X Y cW 1 1 WW PX 1 R SQ PY X V V P V ...
View Full Document

This note was uploaded on 02/11/2012 for the course EEE 352 taught by Professor Ferry during the Spring '08 term at ASU.

Ask a homework question - tutors are online