NCchrisSU All American 1024 Posts user info edit post |
Three FM radio stations (A, B, and C) are competing for customers in the same market area. Through an aggressive advertising campaign, station A is capturing 11% of station B's customers and 5% of station C's customers each month, while losing only 2% of its customers to B and 2% to C each month. And 1% of B's customers switch to C and 1% of C's customers switch to B each month. This gives you all the information you need to set up the Markov chain for this situation.
Initially A and B each have 30% of the listeners and C has 40%
What percentage will radio station A have in the long run?
What percentage will radio station B have in the long run?
What percentage will radio station C have in the long run?
thats all but like I said... I do have the answers since the webassign is past due its .6396 for A, ..1261 for B and .2342 for C.... but I dont know how to get the answer and the final is tomorrow and I know this is going to be on them 12/8/2005 12:56:23 AM |
clalias All American 1580 Posts user info edit post |
To set the problem up; Let A1,B1,C1 be the #of customers and, Ao,Bo,Co be the initial number of customers.
From the first statement Quote : | "station A is capturing 11% of station B's customers and 5% of station C's customers each month, while losing only 2% of its customers to B and 2% to C each month." |
Because A is gaining 11%of B and 5%of C we get ( the __ means unknown placeholder) A1= __Ao+0.11Bo+0.05Co, but we know that A is losing a total of 4%of it's customers so we know,
A1=0.96Ao+0.11Bo+0.05Co
From the same quote above we know that B and C are getting 2% of A, so we write
B1=0.02Ao+__Bo+__Co C1=0.02Ao+__Bo+__Co
Now from the last bit of infoQuote : | "And 1% of B's customers switch to C and 1% of C's customers switch to B each month." |
We get B1=0.02Ao +__Bo +0.01Co C1=0.02Ao+0.01Bo+__Co
We still need to get the last two pieces, so Remember the Columns must add to 1.
A1=0.96Ao+0.11Bo+0.05Co B1=0.02Ao +__Bo +0.01Co C1=0.02Ao+0.01Bo+__Co
So we must have, A1=0.96Ao+0.11Bo+0.05Co B1=0.02Ao +0.88Bo +0.01Co C1=0.02Ao+0.01Bo+0.94Co
Now this sets up the Markov process. I have never taken Finite math so I don't know what tools you have to get A_infinity. I would use the eigen decomposition.
Here is the Matlab output to prove this is right...
>> A=[0.96,0.11,0.05;0.02,0.88,0.01;0.02,0.01,0.94]
A =
0.9600 0.1100 0.0500 0.0200 0.8800 0.0100 0.0200 0.0100 0.9400
>> A^10000
ans =
0.6396 0.6396 0.6396 0.1261 0.1261 0.1261 0.2342 0.2342 0.2342
>>12/8/2005 2:19:25 AM |