2005_03_01_archive

Medicare debt = 7x social security debt

This MSNBC article shows the magnitude of the problem, and reminds me

of Design Challenge II back in January, when we had to analyze the

healthcare system in the US and make recommendations. Tough cookie

that one. It will probably too late when we finally confront it.

But on the plus side, now that ERBA HW4 is finally finished, I can

turn my attention to more interesting, not to mention infinitely more

useful, items.

Posted by Yoav at 20:27 1 comments Links to this post

Ice Harvesting

Today's Disruptive Technologies class covered the history of the ice

harvesting industry from start to today's status. Professor Utterback

was marvelous as usual. It's fascinating to have a lecturer with such

breadth and depth of knowledge in so many industries. He keeps saying

he's not a historian, but his knowledge is truly impressive...

We're all still working on the ERBA homework question below -- it

sucks... Stay tuned ;)

Posted by Yoav at 16:04 0 comments Links to this post

ERBA Homework #4...

... is driving the class crazy. I've solved 3 out of the 4 problems

without much difficulty, but the fourth problem is annoying. Here it

is below, and if you have any advice, please do share... ;)

4) Suppose that the time to failure T (years) of a certain item is an

exponential random variable with density f(t/lambda) =

lambda*exp(-lambda*t), t>0.

From prior experience, we are led to believe that lambda is a value of

an exponential random variable with density pi(lambda) =

2*exp(-2*lambda).

a) What are the aleatory and epistemic models in this case?

b) If a sample of three observations is (1; 3; 5), plot the

(posterior) 5^th , 50^th, and 95^th percentile (aleatory) reliability

curves as functions of time. What is the range of values of the

reliability for a period of one year?

c) What are the prior and posterior mean values of

[C:%5CDOCUME%7E1%5CYoav%5CLOCALS%7E1%5CTemp%5Cmsohtml1%5C01%5Cclip_ima

d) Find the point estimate for

[C:%5CDOCUME%7E1%5CYoav%5CLOCALS%7E1%5CTemp%5Cmsohtml1%5C01%5Cclip_ima