*** empty log message ***

This commit is contained in:
George Weigt 2006-10-26 14:51:30 -07:00
parent 0ec01a2ed1
commit 3e86c5571c

50
131.tex Normal file
View file

@ -0,0 +1,50 @@
3. Let random variable $X$ be the time to leakage failure in
a certain type of dry-cell battery with probability
density function
$$f(x)=\cases{
\displaystyle{
{\beta\over\sigma^\beta}(x-\tau)^{\beta-1}e^{-\left({x-\tau\over\sigma}\right)^\beta}
},
&for $x\in(\tau,\infty)$\cr
\cr
0,&otherwise
}$$
\bigskip
{\bf (a) Find the mean and variance of $X$.}
$$\mu=\int_\tau^\infty x{\beta\over\sigma^\beta}(x-\tau)^{\beta-1}e^{-(x-\tau)^\beta/\sigma^\beta}\,dx$$
Use integration by parts.
$$
u=x,\qquad du=dx,\qquad
v=-e^{-(x-\tau)^\beta/\sigma^\beta},\qquad
dv={\beta\over\sigma^\beta}(x-\tau)^{\beta-1}e^{-(x-\tau)^\beta/\sigma^\beta}\,dx
$$
$$
\int u\,dv=uv-\int v\,du
=-xe^{-(x-\tau)^\beta/\sigma^\beta}
+\int e^{-(x-\tau)^\beta/\sigma^\beta}\,dx
$$
Can't figure out the rest. Is the mean $\tau$?
\bigskip
{\bf (b) Find the distribution function of $X$.}
\bigskip
$$\int_\tau^t{\beta\over\sigma^\beta}(x-\tau)^{\beta-1}e^{-(x-\tau)^\beta/\sigma^\beta}
=-e^{-(x-\tau)^\beta/\sigma^\beta}\bigg|_\tau^t
=1-e^{-(t-\tau)^\beta/\sigma^\beta}
$$
$$F(t)=\cases{
0,&$t<\tau$\cr
1-e^{-(t-\tau)^\beta/\sigma^\beta},&$\tau\le t$\cr
}$$
\bigskip
{\bf (c) What is the distribution of $X$ when $\tau=0$, $\beta=1$?}
$$F(t)\bigg|_{\tau=0,\,\beta=1}=\cases{
0,&$t<1$\cr
1-e^{-t/\sigma},&$1\le t$\cr
}$$
\end