(New page: ==7.10 QE 2005 August== 1. (30 Points) Assume that <math>\mathbf{X}</math> is a binomial distributed random variable with probability mass function (pmf) given by <math>p_{n}\left(k\rig...)
 
 
(7 intermediate revisions by 2 users not shown)
Line 1: Line 1:
==7.10 QE 2005 August==
+
==7.10 [[ECE_PhD_Qualifying_Exams|QE]] 2005 August==
  
1. (30 Points)
+
'''1. (30 Points)'''
  
Assume that <math>\mathbf{X}</math>  is a binomial distributed random variable with probability mass function (pmf) given by <math>p_{n}\left(k\right)=\left(\begin{array}{c}
+
Assume that <math class="inline">\mathbf{X}</math>  is a binomial distributed random variable with probability mass function (pmf) given by <math class="inline">p_{n}\left(k\right)=\left(\begin{array}{c}
 
n\\
 
n\\
 
k
 
k
\end{array}\right)p^{k}\left(1-p\right)^{n-k}\;,\qquad k=0,1,2,\cdots,n,</math><math> where <math>0<p<1</math> .
+
\end{array}\right)p^{k}\left(1-p\right)^{n-k}\;,\qquad k=0,1,2,\cdots,n</math> where <math class="inline">0<p<1</math> .
  
(a)
+
'''(a)'''
  
Find the characteristic function of <math>\mathbf{X}</math> . (You must show how you derive the characteristic function.)
+
Find the characteristic function of <math class="inline">\mathbf{X}</math> . (You must show how you derive the characteristic function.)
  
<math>\Phi_{\mathbf{X}}\left(\omega\right)=E\left[e^{i\omega\mathbf{X}}\right]=\sum_{k=0}^{n}e^{i\omega k}\cdot\left(\begin{array}{c}
+
<math class="inline">\Phi_{\mathbf{X}}\left(\omega\right)=E\left[e^{i\omega\mathbf{X}}\right]=\sum_{k=0}^{n}e^{i\omega k}\cdot\left(\begin{array}{c}
 
n\\
 
n\\
 
k
 
k
Line 18: Line 18:
 
n\\
 
n\\
 
k
 
k
\end{array}\right)\left(p\cdot e^{i\omega}\right)^{k}\left(1-p\right)^{n-k}</math><math>=\left(p\cdot e^{i\omega}+1-p\right)^{n}.</math>  
+
\end{array}\right)\left(p\cdot e^{i\omega}\right)^{k}\left(1-p\right)^{n-k}</math><math class="inline">=\left(p\cdot e^{i\omega}+1-p\right)^{n}.</math>  
  
(b)  
+
'''(b)'''
  
Compute the standard deviation of <math>\mathbf{X}</math> .
+
Compute the standard deviation of <math class="inline">\mathbf{X}</math> .
  
<math>E\left[\mathbf{X}\right]=\frac{d}{di\omega}\Phi_{\mathbf{X}}\left(\omega\right)\Bigl|_{i\omega=0}=n\left(pe^{i\omega}+1-p\right)^{n-1}\cdot pe^{i\omega}\Bigl|_{i\omega=0}=np.</math>  
+
<math class="inline">E\left[\mathbf{X}\right]=\frac{d}{di\omega}\Phi_{\mathbf{X}}\left(\omega\right)\Bigl|_{i\omega=0}=n\left(pe^{i\omega}+1-p\right)^{n-1}\cdot pe^{i\omega}\Bigl|_{i\omega=0}=np.</math>  
  
<math>E\left[\mathbf{X}^{2}\right]=\frac{d^{2}}{d\left(i\omega\right)^{2}}\Phi_{\mathbf{X}}\left(\omega\right)\Bigl|_{i\omega=0}=\frac{d}{di\omega}npe^{i\omega}\left(pe^{i\omega}+1-p\right)^{n-1}\Bigl|_{i\omega=0}</math><math>=np\left[\frac{d}{di\omega}e^{i\omega}\left(pe^{i\omega}+1-p\right)^{n-1}\Bigl|_{i\omega=0}\right]</math><math>=np\left[e^{i\omega}\left(pe^{i\omega}+1-p\right)^{n-1}+e^{i\omega}(n-1)\left(pe^{i\omega}+1-p\right)^{n-2}\cdot pe^{i\omega}\Bigl|_{i\omega=0}\right]</math><math>=np\left[1+\left(n-1\right)p\right]=np+n\left(n-1\right)p^{2}.</math>  
+
<math class="inline">E\left[\mathbf{X}^{2}\right]=\frac{d^{2}}{d\left(i\omega\right)^{2}}\Phi_{\mathbf{X}}\left(\omega\right)\Bigl|_{i\omega=0}=\frac{d}{di\omega}npe^{i\omega}\left(pe^{i\omega}+1-p\right)^{n-1}\Bigl|_{i\omega=0}</math><math class="inline">=np\left[\frac{d}{di\omega}e^{i\omega}\left(pe^{i\omega}+1-p\right)^{n-1}\Bigl|_{i\omega=0}\right]</math><math class="inline">=np\left[e^{i\omega}\left(pe^{i\omega}+1-p\right)^{n-1}+e^{i\omega}(n-1)\left(pe^{i\omega}+1-p\right)^{n-2}\cdot pe^{i\omega}\Bigl|_{i\omega=0}\right]</math><math class="inline">=np\left[1+\left(n-1\right)p\right]=np+n\left(n-1\right)p^{2}.</math>  
  
<math>Var\left[\mathbf{X}\right]=E\left[\mathbf{X}^{2}\right]-\left(E\left[\mathbf{X}\right]\right)^{2}=np+n\left(n-1\right)p^{2}-\left(np\right)^{2}=np+n^{2}p^{2}-np^{2}-n^{2}p^{2}=np\left(1-p\right).</math>  
+
<math class="inline">Var\left[\mathbf{X}\right]=E\left[\mathbf{X}^{2}\right]-\left(E\left[\mathbf{X}\right]\right)^{2}=np+n\left(n-1\right)p^{2}-\left(np\right)^{2}=np+n^{2}p^{2}-np^{2}-n^{2}p^{2}=np\left(1-p\right).</math>  
  
<math>\therefore\sigma_{\mathbf{X}}=\sqrt{np\left(1-p\right)}.</math>  
+
<math class="inline">\therefore\sigma_{\mathbf{X}}=\sqrt{np\left(1-p\right)}.</math>  
  
(c)
+
'''(c)'''
  
Find the value or values of <math>k</math>  for which <math>p_{n}\left(k\right)</math>  is maximum, and express the answer in terms of <math>p</math>  and <math>n</math> . Give the most complete answer to this question that you can.
+
Find the value or values of <math class="inline">k</math>  for which <math class="inline">p_{n}\left(k\right)</math>  is maximum, and express the answer in terms of <math class="inline">p</math>  and <math class="inline">n</math> . Give the most complete answer to this question that you can.
  
<math>\frac{p_{n}\left(k\right)}{p_{n}\left(k-1\right)}=\frac{\left(\begin{array}{c}
+
<math class="inline">\frac{p_{n}\left(k\right)}{p_{n}\left(k-1\right)}=\frac{\left(\begin{array}{c}
 
n\\
 
n\\
 
k
 
k
Line 42: Line 42:
 
n\\
 
n\\
 
k-1
 
k-1
\end{array}\right)p^{k-1}\left(1-p\right)^{n-k+1}}=\frac{\frac{n!}{\left(n-k\right)!k!}p^{k}\left(1-p\right)^{n-k}}{\frac{n!}{\left(n-k+1\right)!\left(k-1\right)!}p^{k-1}\left(1-p\right)^{n-k+1}}</math><math>=\frac{n!}{\left(n-k\right)!k!}\cdot\frac{\left(n-k+1\right)!\left(k-1\right)!}{n!}\cdot\frac{p}{1-p}=\frac{n-k+1}{k}\cdot\frac{p}{1-p}.</math>
+
\end{array}\right)p^{k-1}\left(1-p\right)^{n-k+1}}=\frac{\frac{n!}{\left(n-k\right)!k!}p^{k}\left(1-p\right)^{n-k}}{\frac{n!}{\left(n-k+1\right)!\left(k-1\right)!}p^{k-1}\left(1-p\right)^{n-k+1}}</math><math class="inline">=\frac{n!}{\left(n-k\right)!k!}\cdot\frac{\left(n-k+1\right)!\left(k-1\right)!}{n!}\cdot\frac{p}{1-p}=\frac{n-k+1}{k}\cdot\frac{p}{1-p}.</math>
  
<math>\frac{p_{n}\left(k\right)}{p_{n}\left(k-1\right)}</math>  is monotonically descreasing function of <math>k</math> . Hence, <math>p_{n}\left(k\right)</math>  is maximized by the largest <math>k</math>  such that <math>\frac{p_{n}\left(k\right)}{p_{n}\left(k-1\right)}\geq1</math> .<math>\left(n-k+1\right)p\geq k\left(1-p\right)\Longrightarrow np-kp+p\geq k-kp\Longrightarrow\left(n+1\right)p\geq k</math>. <math>k</math>  should be the integer. Thus, <math>k=\left\lfloor \left(n+1\right)p\right\rfloor</math>  maximize <math>p_{n}\left(k\right)</math> . If <math>p_{n}\left(k\right)=p_{n}\left(k-1\right)</math> , both <math>k=\left(n+1\right)p</math>  and <math>k=\left(n+1\right)p-1</math>  maximize <math>p_{n}\left(k\right)</math> .
+
<math class="inline">\frac{p_{n}\left(k\right)}{p_{n}\left(k-1\right)}</math>  is monotonically descreasing function of <math class="inline">k</math> . Hence, <math class="inline">p_{n}\left(k\right)</math>  is maximized by the largest <math class="inline">k</math>  such that <math class="inline">\frac{p_{n}\left(k\right)}{p_{n}\left(k-1\right)}\geq1</math> .<math class="inline">\left(n-k+1\right)p\geq k\left(1-p\right)\Longrightarrow np-kp+p\geq k-kp\Longrightarrow\left(n+1\right)p\geq k</math>. <math class="inline">k</math>  should be the integer. Thus, <math class="inline">k=\left\lfloor \left(n+1\right)p\right\rfloor</math>  maximize <math class="inline">p_{n}\left(k\right)</math> . If <math class="inline">p_{n}\left(k\right)=p_{n}\left(k-1\right)</math> , both <math class="inline">k=\left(n+1\right)p</math>  and <math class="inline">k=\left(n+1\right)p-1</math>  maximize <math class="inline">p_{n}\left(k\right)</math> .
  
2. (30 Points)
+
'''2. (30 Points)'''
  
Let <math>\mathbf{X}_{1},\mathbf{X}_{2},\cdots,\mathbf{X}_{n},\cdots</math>  be a sequence of binomially distributed random variables, with <math>\mathbf{X}_{n}</math>  having probability mass function <math>p_{n}\left(k\right)=\left(\begin{array}{c}
+
Let <math class="inline">\mathbf{X}_{1},\mathbf{X}_{2},\cdots,\mathbf{X}_{n},\cdots</math>  be a sequence of binomially distributed random variables, with <math class="inline">\mathbf{X}_{n}</math>  having probability mass function <math class="inline">p_{n}\left(k\right)=\left(\begin{array}{c}
 
n\\
 
n\\
 
k
 
k
\end{array}\right)p_{n}^{k}\left(1-p_{n}\right)^{n-k}\;,\qquad k=0,1,2,\cdots,n,</math> where <math>0<p_{n}<1</math>  for all <math>n=1,2,3,\cdots</math> . Show that if <math>np_{n}\rightarrow\lambda\text{ as }n\rightarrow\infty,</math> then the random sequence <math>\mathbf{X}_{1},\mathbf{X}_{2},\cdots,\mathbf{X}_{n},\cdots</math>  converges in distribution to a Poisson random variable having mean <math>\lambda</math> .
+
\end{array}\right)p_{n}^{k}\left(1-p_{n}\right)^{n-k}\;,\qquad k=0,1,2,\cdots,n,</math> where <math class="inline">0<p_{n}<1</math>  for all <math class="inline">n=1,2,3,\cdots</math> . Show that if <math class="inline">np_{n}\rightarrow\lambda\text{ as }n\rightarrow\infty,</math> then the random sequence <math class="inline">\mathbf{X}_{1},\mathbf{X}_{2},\cdots,\mathbf{X}_{n},\cdots</math>  converges in distribution to a Poisson random variable having mean <math class="inline">\lambda</math> .
  
<math>\Phi_{\mathbf{X}_{n}}\left(\omega\right)=E\left[e^{i\omega\mathbf{X}_{n}}\right]=\sum_{k=0}^{n}e^{i\omega k}\left(\begin{array}{c}
+
<math class="inline">\Phi_{\mathbf{X}_{n}}\left(\omega\right)=E\left[e^{i\omega\mathbf{X}_{n}}\right]=\sum_{k=0}^{n}e^{i\omega k}\left(\begin{array}{c}
 
n\\
 
n\\
 
k
 
k
Line 59: Line 59:
 
n\\
 
n\\
 
k
 
k
\end{array}\right)\left(e^{i\omega}p_{n}\right)^{k}\left(1-p_{n}\right)^{n-k}</math><math>=\left(e^{i\omega}p_{n}+1-p_{n}\right)^{n}=\left(1-p_{n}\left(e^{i\omega}-1\right)\right)^{n}.</math>
+
\end{array}\right)\left(e^{i\omega}p_{n}\right)^{k}\left(1-p_{n}\right)^{n-k}</math><math class="inline">=\left(e^{i\omega}p_{n}+1-p_{n}\right)^{n}=\left(1-p_{n}\left(e^{i\omega}-1\right)\right)^{n}.</math>
  
Since <math>np_{n}\rightarrow\lambda</math>  as <math>n\rightarrow\infty</math> , <math>p_{n}\rightarrow\lambda/n</math> . <math>\lim_{n\rightarrow\infty}\Phi_{\mathbf{X}_{n}}\left(\omega\right)=\lim_{n\rightarrow\infty}\left(1-p_{n}\left(e^{i\omega}-1\right)\right)^{n}=\lim_{n\rightarrow\infty}\left(1-\frac{\lambda}{n}\left(e^{i\omega}-1\right)\right)^{n}=e^{\lambda\left(e^{i\omega}-1\right)}.</math>  
+
Since <math class="inline">np_{n}\rightarrow\lambda</math>  as <math class="inline">n\rightarrow\infty</math> , <math class="inline">p_{n}\rightarrow\lambda/n</math> . <math class="inline">\lim_{n\rightarrow\infty}\Phi_{\mathbf{X}_{n}}\left(\omega\right)=\lim_{n\rightarrow\infty}\left(1-p_{n}\left(e^{i\omega}-1\right)\right)^{n}=\lim_{n\rightarrow\infty}\left(1-\frac{\lambda}{n}\left(e^{i\omega}-1\right)\right)^{n}=e^{\lambda\left(e^{i\omega}-1\right)}.</math>  
  
<math>\because\lim_{n\rightarrow\infty}\left(1+\frac{x}{n}\right)^{n}=e^{x}.</math>  
+
<math class="inline">\because\lim_{n\rightarrow\infty}\left(1+\frac{x}{n}\right)^{n}=e^{x}.</math>  
  
<math>e^{\lambda\left(e^{i\omega}-1\right)}</math>  is the characteristic function of a Poisson random variable with mean <math>\lambda</math> . Thus, as <math>n\rightarrow\infty</math> , <math>\mathbf{X}_{n}</math>  converges in distribution to a Poisson random variable with mean <math>\lambda</math> .
+
<math class="inline">e^{\lambda\left(e^{i\omega}-1\right)}</math>  is the characteristic function of a Poisson random variable with mean <math class="inline">\lambda</math> . Thus, as <math class="inline">n\rightarrow\infty</math> , <math class="inline">\mathbf{X}_{n}</math>  converges in distribution to a Poisson random variable with mean <math class="inline">\lambda</math> .
  
3. (40 Points)
+
'''3. (40 Points)'''
  
Consider a homogeneous Poisson point process with rate <math>\lambda</math>  and points (event occurrence times) <math>\mathbf{T}_{1},\mathbf{T}_{2},\cdots,\mathbf{T}_{n},\cdots</math> .
+
Consider a homogeneous Poisson point process with rate <math class="inline">\lambda</math>  and points (event occurrence times) <math class="inline">\mathbf{T}_{1},\mathbf{T}_{2},\cdots,\mathbf{T}_{n},\cdots</math> .
  
(a)
+
'''(a)'''
  
Derive the pdf <math>f_{k}\left(t\right)</math>  of the <math>k</math> -th point <math>\mathbf{T}_{k}</math>  for arbitrary <math>k</math> .
+
Derive the pdf <math class="inline">f_{k}\left(t\right)</math>  of the <math class="inline">k</math> -th point <math class="inline">\mathbf{T}_{k}</math>  for arbitrary <math class="inline">k</math> .
  
 
• The cdf is  
 
• The cdf is  
<math>F_{\mathbf{T}_{k}}\left(t\right)=P\left(\mathbf{T}_{k}\leq t\right)=P\left(\text{at least }k\text{ points within }t\right)=\sum_{j=k}^{\infty}\frac{e^{-\lambda t}\cdot\left(\lambda t\right)^{j}}{j!}</math><math>=1-P\left(\mathbf{T}_{k}>t\right)=1-P\left(\text{less than }k\text{ points within }t\right)=1-\sum_{j=0}^{k-1}\frac{e^{-\lambda t}\cdot\left(\lambda t\right)^{j}}{j!}.</math>
+
<math class="inline">F_{\mathbf{T}_{k}}\left(t\right)=P\left(\mathbf{T}_{k}\leq t\right)=P\left(\text{at least }k\text{ points within }t\right)=\sum_{j=k}^{\infty}\frac{e^{-\lambda t}\cdot\left(\lambda t\right)^{j}}{j!}</math><math class="inline">=1-P\left(\mathbf{T}_{k}>t\right)=1-P\left(\text{less than }k\text{ points within }t\right)=1-\sum_{j=0}^{k-1}\frac{e^{-\lambda t}\cdot\left(\lambda t\right)^{j}}{j!}.</math>
  
 
• The pdf by differentiating the cdf  
 
• The pdf by differentiating the cdf  
<math>isf_{\mathbf{T}_{k}}\left(t\right)=\frac{dF_{\mathbf{T}_{k}}}{dt}=-\sum_{j=0}^{k-1}\frac{\left(-\lambda\right)e^{-\lambda t}\cdot\left(\lambda t\right)^{j}}{j!}-\sum_{j=0}^{k-1}\frac{e^{-\lambda t}\cdot j\left(\lambda t\right)^{j-1}\cdot\lambda}{j!}</math><math>=\lambda e^{-\lambda t}\sum_{j=0}^{k-1}\frac{\left(\lambda t\right)^{j}}{j!}-\lambda e^{-\lambda t}\sum_{j=1}^{k-1}\frac{j\left(\lambda t\right)^{j-1}}{j!}=\lambda e^{-\lambda t}\sum_{j=0}^{k-1}\frac{\left(\lambda t\right)^{j}}{j!}-\lambda e^{-\lambda t}\sum_{j=1}^{k-1}\frac{\left(\lambda t\right)^{j-1}}{\left(j-1\right)!}</math><math>=\lambda e^{-\lambda t}\sum_{j=0}^{k-1}\frac{\left(\lambda t\right)^{j}}{j!}-\lambda e^{-\lambda t}\sum_{j=0}^{k-2}\frac{\left(\lambda t\right)^{j}}{j!}=\lambda e^{-\lambda t}\frac{\left(\lambda t\right)^{k-1}}{\left(k-1\right)!}.</math>  
+
<math class="inline">isf_{\mathbf{T}_{k}}\left(t\right)=\frac{dF_{\mathbf{T}_{k}}}{dt}=-\sum_{j=0}^{k-1}\frac{\left(-\lambda\right)e^{-\lambda t}\cdot\left(\lambda t\right)^{j}}{j!}-\sum_{j=0}^{k-1}\frac{e^{-\lambda t}\cdot j\left(\lambda t\right)^{j-1}\cdot\lambda}{j!}</math><math class="inline">=\lambda e^{-\lambda t}\sum_{j=0}^{k-1}\frac{\left(\lambda t\right)^{j}}{j!}-\lambda e^{-\lambda t}\sum_{j=1}^{k-1}\frac{j\left(\lambda t\right)^{j-1}}{j!}=\lambda e^{-\lambda t}\sum_{j=0}^{k-1}\frac{\left(\lambda t\right)^{j}}{j!}-\lambda e^{-\lambda t}\sum_{j=1}^{k-1}\frac{\left(\lambda t\right)^{j-1}}{\left(j-1\right)!}</math><math class="inline">=\lambda e^{-\lambda t}\sum_{j=0}^{k-1}\frac{\left(\lambda t\right)^{j}}{j!}-\lambda e^{-\lambda t}\sum_{j=0}^{k-2}\frac{\left(\lambda t\right)^{j}}{j!}=\lambda e^{-\lambda t}\frac{\left(\lambda t\right)^{k-1}}{\left(k-1\right)!}.</math>  
  
 
– This is Erlang distribution.
 
– This is Erlang distribution.
  
(b)
+
'''(b)'''
  
What kind of distribution does <math>\mathbf{T}_{1}</math>  have?
+
What kind of distribution does <math class="inline">\mathbf{T}_{1}</math>  have?
  
• If <math>k=1</math> , then <math>f_{\mathbf{T}_{1}}\left(t\right)=\lambda e^{-\lambda t}</math> . Thus <math>\mathbf{T}_{1}</math>  is a exponential random variable with parameter <math>\lambda</math> .
+
• If <math class="inline">k=1</math> , then <math class="inline">f_{\mathbf{T}_{1}}\left(t\right)=\lambda e^{-\lambda t}</math> . Thus <math class="inline">\mathbf{T}_{1}</math>  is a exponential random variable with parameter <math class="inline">\lambda</math> .
  
(c)
+
'''(c)'''
  
What is the conditional pdf of <math>\mathbf{T}_{k}</math>  given <math>\mathbf{T}_{k-1}=t_{0}</math> , where <math>t_{0}>0</math> ? (You can give the answer without derivation if you know it.)
+
What is the conditional pdf of <math class="inline">\mathbf{T}_{k}</math>  given <math class="inline">\mathbf{T}_{k-1}=t_{0}</math> , where <math class="inline">t_{0}>0</math> ? (You can give the answer without derivation if you know it.)
  
 
• The conditional cdf is
 
• The conditional cdf is
  
<math>F_{\mathbf{T}_{k}}\left(t_{k}|\mathbf{T}_{k-1}=t_{0}\right)=P\left(\mathbf{T}_{k}\leq t_{k}|\mathbf{T}_{k-1}=t_{0}\right)=P\left(N\left(t_{k},t_{0}\right)\geq1\right)</math><math>=1-P\left(N\left(t_{k},t_{0}\right)=0\right)=1-e^{-\lambda\left(t_{k}-t_{0}\right)}.</math>  
+
<math class="inline">F_{\mathbf{T}_{k}}\left(t_{k}|\mathbf{T}_{k-1}=t_{0}\right)=P\left(\mathbf{T}_{k}\leq t_{k}|\mathbf{T}_{k-1}=t_{0}\right)=P\left(N\left(t_{k},t_{0}\right)\geq1\right)</math><math class="inline">=1-P\left(N\left(t_{k},t_{0}\right)=0\right)=1-e^{-\lambda\left(t_{k}-t_{0}\right)}.</math>  
  
 
• The conditional pdf by differentiating the conditional cdf is
 
• The conditional pdf by differentiating the conditional cdf is
  
<math>\therefore f_{\mathbf{T}_{k}}\left(t_{k}|\mathbf{T}_{k-1}=t_{0}\right)=\lambda e^{-\lambda\left(t_{k}-t_{0}\right)}.</math>  
+
<math class="inline">\therefore f_{\mathbf{T}_{k}}\left(t_{k}|\mathbf{T}_{k-1}=t_{0}\right)=\lambda e^{-\lambda\left(t_{k}-t_{0}\right)}.</math>  
  
(d)
+
'''(d)'''
  
Suppose you have a random number generator that produces independent, identically distributed (i.i.d. ) random variables <math>\mathbf{X}_{1},\mathbf{X}_{2},\cdots,\mathbf{X}_{n},\cdots</math>  that are uniformaly distributed on the interval <math>\left(0,1\right)</math> . Explain how you could use these to simulate the Poisson points <math>\mathbf{T}_{1},\mathbf{T}_{2},\cdots,\mathbf{T}_{n},\cdots</math>  describe above. Provide as complete an explanation as possible.
+
Suppose you have a random number generator that produces independent, identically distributed (i.i.d. ) random variables <math class="inline">\mathbf{X}_{1},\mathbf{X}_{2},\cdots,\mathbf{X}_{n},\cdots</math>  that are uniformaly distributed on the interval <math class="inline">\left(0,1\right)</math> . Explain how you could use these to simulate the Poisson points <math class="inline">\mathbf{T}_{1},\mathbf{T}_{2},\cdots,\mathbf{T}_{n},\cdots</math>  describe above. Provide as complete an explanation as possible.
  
• This problem is similar to QE 2003 August [CS1QE2003August] Problem 2.  
+
• This problem is similar to [[ECE600 QE 2003 August|QE 2003 August]] Problem 2.  
  
 
----
 
----
 
[[ECE600|Back to ECE600]]
 
[[ECE600|Back to ECE600]]
  
[[ECE 600 QE|Back to ECE 600 QE]]
+
[[ECE 600 QE|Back to my ECE 600 QE page]]
 +
 
 +
[[ECE_PhD_Qualifying_Exams|Back to the general ECE PHD QE page]] (for problem discussion)

Latest revision as of 07:30, 27 June 2012

7.10 QE 2005 August

1. (30 Points)

Assume that $ \mathbf{X} $ is a binomial distributed random variable with probability mass function (pmf) given by $ p_{n}\left(k\right)=\left(\begin{array}{c} n\\ k \end{array}\right)p^{k}\left(1-p\right)^{n-k}\;,\qquad k=0,1,2,\cdots,n $ where $ 0<p<1 $ .

(a)

Find the characteristic function of $ \mathbf{X} $ . (You must show how you derive the characteristic function.)

$ \Phi_{\mathbf{X}}\left(\omega\right)=E\left[e^{i\omega\mathbf{X}}\right]=\sum_{k=0}^{n}e^{i\omega k}\cdot\left(\begin{array}{c} n\\ k \end{array}\right)p^{k}\left(1-p\right)^{n-k}=\sum_{k=0}^{n}\left(\begin{array}{c} n\\ k \end{array}\right)\left(p\cdot e^{i\omega}\right)^{k}\left(1-p\right)^{n-k} $$ =\left(p\cdot e^{i\omega}+1-p\right)^{n}. $

(b)

Compute the standard deviation of $ \mathbf{X} $ .

$ E\left[\mathbf{X}\right]=\frac{d}{di\omega}\Phi_{\mathbf{X}}\left(\omega\right)\Bigl|_{i\omega=0}=n\left(pe^{i\omega}+1-p\right)^{n-1}\cdot pe^{i\omega}\Bigl|_{i\omega=0}=np. $

$ E\left[\mathbf{X}^{2}\right]=\frac{d^{2}}{d\left(i\omega\right)^{2}}\Phi_{\mathbf{X}}\left(\omega\right)\Bigl|_{i\omega=0}=\frac{d}{di\omega}npe^{i\omega}\left(pe^{i\omega}+1-p\right)^{n-1}\Bigl|_{i\omega=0} $$ =np\left[\frac{d}{di\omega}e^{i\omega}\left(pe^{i\omega}+1-p\right)^{n-1}\Bigl|_{i\omega=0}\right] $$ =np\left[e^{i\omega}\left(pe^{i\omega}+1-p\right)^{n-1}+e^{i\omega}(n-1)\left(pe^{i\omega}+1-p\right)^{n-2}\cdot pe^{i\omega}\Bigl|_{i\omega=0}\right] $$ =np\left[1+\left(n-1\right)p\right]=np+n\left(n-1\right)p^{2}. $

$ Var\left[\mathbf{X}\right]=E\left[\mathbf{X}^{2}\right]-\left(E\left[\mathbf{X}\right]\right)^{2}=np+n\left(n-1\right)p^{2}-\left(np\right)^{2}=np+n^{2}p^{2}-np^{2}-n^{2}p^{2}=np\left(1-p\right). $

$ \therefore\sigma_{\mathbf{X}}=\sqrt{np\left(1-p\right)}. $

(c)

Find the value or values of $ k $ for which $ p_{n}\left(k\right) $ is maximum, and express the answer in terms of $ p $ and $ n $ . Give the most complete answer to this question that you can.

$ \frac{p_{n}\left(k\right)}{p_{n}\left(k-1\right)}=\frac{\left(\begin{array}{c} n\\ k \end{array}\right)p^{k}\left(1-p\right)^{n-k}}{\left(\begin{array}{c} n\\ k-1 \end{array}\right)p^{k-1}\left(1-p\right)^{n-k+1}}=\frac{\frac{n!}{\left(n-k\right)!k!}p^{k}\left(1-p\right)^{n-k}}{\frac{n!}{\left(n-k+1\right)!\left(k-1\right)!}p^{k-1}\left(1-p\right)^{n-k+1}} $$ =\frac{n!}{\left(n-k\right)!k!}\cdot\frac{\left(n-k+1\right)!\left(k-1\right)!}{n!}\cdot\frac{p}{1-p}=\frac{n-k+1}{k}\cdot\frac{p}{1-p}. $

$ \frac{p_{n}\left(k\right)}{p_{n}\left(k-1\right)} $ is monotonically descreasing function of $ k $ . Hence, $ p_{n}\left(k\right) $ is maximized by the largest $ k $ such that $ \frac{p_{n}\left(k\right)}{p_{n}\left(k-1\right)}\geq1 $ .$ \left(n-k+1\right)p\geq k\left(1-p\right)\Longrightarrow np-kp+p\geq k-kp\Longrightarrow\left(n+1\right)p\geq k $. $ k $ should be the integer. Thus, $ k=\left\lfloor \left(n+1\right)p\right\rfloor $ maximize $ p_{n}\left(k\right) $ . If $ p_{n}\left(k\right)=p_{n}\left(k-1\right) $ , both $ k=\left(n+1\right)p $ and $ k=\left(n+1\right)p-1 $ maximize $ p_{n}\left(k\right) $ .

2. (30 Points)

Let $ \mathbf{X}_{1},\mathbf{X}_{2},\cdots,\mathbf{X}_{n},\cdots $ be a sequence of binomially distributed random variables, with $ \mathbf{X}_{n} $ having probability mass function $ p_{n}\left(k\right)=\left(\begin{array}{c} n\\ k \end{array}\right)p_{n}^{k}\left(1-p_{n}\right)^{n-k}\;,\qquad k=0,1,2,\cdots,n, $ where $ 0<p_{n}<1 $ for all $ n=1,2,3,\cdots $ . Show that if $ np_{n}\rightarrow\lambda\text{ as }n\rightarrow\infty, $ then the random sequence $ \mathbf{X}_{1},\mathbf{X}_{2},\cdots,\mathbf{X}_{n},\cdots $ converges in distribution to a Poisson random variable having mean $ \lambda $ .

$ \Phi_{\mathbf{X}_{n}}\left(\omega\right)=E\left[e^{i\omega\mathbf{X}_{n}}\right]=\sum_{k=0}^{n}e^{i\omega k}\left(\begin{array}{c} n\\ k \end{array}\right)p_{n}^{k}\left(1-p_{n}\right)^{n-k}=\sum_{k=0}^{n}\left(\begin{array}{c} n\\ k \end{array}\right)\left(e^{i\omega}p_{n}\right)^{k}\left(1-p_{n}\right)^{n-k} $$ =\left(e^{i\omega}p_{n}+1-p_{n}\right)^{n}=\left(1-p_{n}\left(e^{i\omega}-1\right)\right)^{n}. $

Since $ np_{n}\rightarrow\lambda $ as $ n\rightarrow\infty $ , $ p_{n}\rightarrow\lambda/n $ . $ \lim_{n\rightarrow\infty}\Phi_{\mathbf{X}_{n}}\left(\omega\right)=\lim_{n\rightarrow\infty}\left(1-p_{n}\left(e^{i\omega}-1\right)\right)^{n}=\lim_{n\rightarrow\infty}\left(1-\frac{\lambda}{n}\left(e^{i\omega}-1\right)\right)^{n}=e^{\lambda\left(e^{i\omega}-1\right)}. $

$ \because\lim_{n\rightarrow\infty}\left(1+\frac{x}{n}\right)^{n}=e^{x}. $

$ e^{\lambda\left(e^{i\omega}-1\right)} $ is the characteristic function of a Poisson random variable with mean $ \lambda $ . Thus, as $ n\rightarrow\infty $ , $ \mathbf{X}_{n} $ converges in distribution to a Poisson random variable with mean $ \lambda $ .

3. (40 Points)

Consider a homogeneous Poisson point process with rate $ \lambda $ and points (event occurrence times) $ \mathbf{T}_{1},\mathbf{T}_{2},\cdots,\mathbf{T}_{n},\cdots $ .

(a)

Derive the pdf $ f_{k}\left(t\right) $ of the $ k $ -th point $ \mathbf{T}_{k} $ for arbitrary $ k $ .

• The cdf is $ F_{\mathbf{T}_{k}}\left(t\right)=P\left(\mathbf{T}_{k}\leq t\right)=P\left(\text{at least }k\text{ points within }t\right)=\sum_{j=k}^{\infty}\frac{e^{-\lambda t}\cdot\left(\lambda t\right)^{j}}{j!} $$ =1-P\left(\mathbf{T}_{k}>t\right)=1-P\left(\text{less than }k\text{ points within }t\right)=1-\sum_{j=0}^{k-1}\frac{e^{-\lambda t}\cdot\left(\lambda t\right)^{j}}{j!}. $

• The pdf by differentiating the cdf $ isf_{\mathbf{T}_{k}}\left(t\right)=\frac{dF_{\mathbf{T}_{k}}}{dt}=-\sum_{j=0}^{k-1}\frac{\left(-\lambda\right)e^{-\lambda t}\cdot\left(\lambda t\right)^{j}}{j!}-\sum_{j=0}^{k-1}\frac{e^{-\lambda t}\cdot j\left(\lambda t\right)^{j-1}\cdot\lambda}{j!} $$ =\lambda e^{-\lambda t}\sum_{j=0}^{k-1}\frac{\left(\lambda t\right)^{j}}{j!}-\lambda e^{-\lambda t}\sum_{j=1}^{k-1}\frac{j\left(\lambda t\right)^{j-1}}{j!}=\lambda e^{-\lambda t}\sum_{j=0}^{k-1}\frac{\left(\lambda t\right)^{j}}{j!}-\lambda e^{-\lambda t}\sum_{j=1}^{k-1}\frac{\left(\lambda t\right)^{j-1}}{\left(j-1\right)!} $$ =\lambda e^{-\lambda t}\sum_{j=0}^{k-1}\frac{\left(\lambda t\right)^{j}}{j!}-\lambda e^{-\lambda t}\sum_{j=0}^{k-2}\frac{\left(\lambda t\right)^{j}}{j!}=\lambda e^{-\lambda t}\frac{\left(\lambda t\right)^{k-1}}{\left(k-1\right)!}. $

– This is Erlang distribution.

(b)

What kind of distribution does $ \mathbf{T}_{1} $ have?

• If $ k=1 $ , then $ f_{\mathbf{T}_{1}}\left(t\right)=\lambda e^{-\lambda t} $ . Thus $ \mathbf{T}_{1} $ is a exponential random variable with parameter $ \lambda $ .

(c)

What is the conditional pdf of $ \mathbf{T}_{k} $ given $ \mathbf{T}_{k-1}=t_{0} $ , where $ t_{0}>0 $ ? (You can give the answer without derivation if you know it.)

• The conditional cdf is

$ F_{\mathbf{T}_{k}}\left(t_{k}|\mathbf{T}_{k-1}=t_{0}\right)=P\left(\mathbf{T}_{k}\leq t_{k}|\mathbf{T}_{k-1}=t_{0}\right)=P\left(N\left(t_{k},t_{0}\right)\geq1\right) $$ =1-P\left(N\left(t_{k},t_{0}\right)=0\right)=1-e^{-\lambda\left(t_{k}-t_{0}\right)}. $

• The conditional pdf by differentiating the conditional cdf is

$ \therefore f_{\mathbf{T}_{k}}\left(t_{k}|\mathbf{T}_{k-1}=t_{0}\right)=\lambda e^{-\lambda\left(t_{k}-t_{0}\right)}. $

(d)

Suppose you have a random number generator that produces independent, identically distributed (i.i.d. ) random variables $ \mathbf{X}_{1},\mathbf{X}_{2},\cdots,\mathbf{X}_{n},\cdots $ that are uniformaly distributed on the interval $ \left(0,1\right) $ . Explain how you could use these to simulate the Poisson points $ \mathbf{T}_{1},\mathbf{T}_{2},\cdots,\mathbf{T}_{n},\cdots $ describe above. Provide as complete an explanation as possible.

• This problem is similar to QE 2003 August Problem 2.


Back to ECE600

Back to my ECE 600 QE page

Back to the general ECE PHD QE page (for problem discussion)

Alumni Liaison

Basic linear algebra uncovers and clarifies very important geometry and algebra.

Dr. Paul Garrett