Let's define a series S , with 0 t h element a 0 as:
S ( a 0 ) = { a 1 , a 2 , a 3 , a 4 , . . . }
Where a n = φ a n − 1 + φ 1 o r a n = φ a n − 1 − φ 2 1 with equal probability.
If a 0 = 3 1 , the expected value of a 5 1 can be expressed as:
d e a b c − f
Find: a + b + c + d + e + f
Details and Assumptions:
This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try
refreshing the page, (b) enabling javascript if it is disabled on your browser and,
finally, (c)
loading the
non-javascript version of this page
. We're sorry about the hassle.
According to the question;
a n = φ a n − 1 + φ 1 o r a n = φ a n − 1 − φ 2 1
with equal probabilities.
This means if A n is the expected value then
A n = 2 1 ( φ A n − 1 + φ 1 ) + 2 1 ( φ A n − 1 − φ 2 1 )
⇒ A n = 2 1 ( φ A n − 1 + φ 1 + φ A n − 1 − φ 2 1 )
⇒ A n = 2 1 ( A n − 1 ( φ + φ 1 ) + ( φ 1 − φ 2 1 ) )
Putting the value of φ = 2 5 + 1 and φ 1 = 2 5 − 1 we get ,
⇒ A n = 2 1 ( A n − 1 5 + 5 − 2 )
⇒ A n = 2 1 ( 5 ( A n − 1 + 1 ) − 2 )
⇒ A n = 2 1 ( 5 ( A n − 1 + 1 ) ) − 1
⇒ A n + 1 = 2 5 ( A n − 1 + 1 )
Now let A n + 1 = T n with T 0 = 3 2
⇒ T n = 2 5 T n − 1
Now a GP if formed with common ratio 2 5
This gives us T 5 1 = T 0 ( 2 5 ) 5 1
and A 5 1 = T 5 1 − 1 = ( 3 2 ) ( 2 5 ) 5 1 − 1
⇒ A 5 1 = 2 4 6 5 2 5 5 − 1
⇒ a + b + c + d + e + f = 5 + 2 5 + 5 + 2 + 4 6 + 1 = 8 4
Nice solution using the value of ϕ beforehand to make calculation easier! (+1)
Problem Loading...
Note Loading...
Set Loading...
Okay, consider this:
Define operation O 1 ( a n − 1 ) = α a n − 1 + β and
O 2 ( a n − 1 ) = α a n − 1 − α β
Observe that O 1 and O 2 are inverse of each other. The series in the question S ( a 0 ) is formed by repeatedly performing these operations(with equal probability) on the terms.
Now, let's say we did this a few times. Some of the operations we made may be O 1 and the rest O 2 . Note that performing O 1 negates the effect of one O 2 , and vice-versa.
So, if O 1 is done p times and O 2 , q times while making 2 m + 1 terms of S ( a 0 ) .
⇒ p + q = 2 m + 1
Now we need only bother about which one of p or q is more. let me illustrate: assume p = 5 , q = 4 . This means that we performed O 1 , 5 times and O 2 , 4 times. Since doing O 2 4 times negates the effect of 4 O 1 s we can see that the net effect after performing the 9 operations is equivalent to performing O 1 once.
Also note that since 2 m + 1 is odd, p and q may differ only by odd numbers.
Hence, to find the expected value, we repeat the following algorithm from p − q = − ( 2 m + 1 ) to p − q = 2 m + 1 :
Find probability P ( p ) of p − q taking given value.
Multiply P with value of a 0 when O 1 is performed on it p times and O 2 is performed q times.
Add to result.
The probability of p being a given value is easy to find and is:
P ( p ) = 2 m + 1 ( 2 m + 1 p )
for example, P ( 2 ) will give probability of p being 2 or alternatively, the probability of q being 2 m − 1 .
Now what's left to do is to find the general term for when O 1 and O 2 are performed n times.
I am writing the formula here without proof, and leaving it as an exercise to the reader to arrive at this formula.
X n = O 1 ( O 1 ( . . n t i m e s ( a 0 ) ) ) = ( a 0 + α − 1 β ) α n − α − 1 β Y n = O 2 ( O 2 ( . . n t i m e s ( a 0 ) ) ) = ( a 0 + α − 1 β ) α n 1 − α − 1 β
After performing the algorithm for finding expected value, the required value( Q ) is:
Q = 2 2 m + 1 1 × [ ∑ i = 1 m P ( m + i ) X 2 i + 1 + ∑ j = 1 m P ( m + j ) Y 2 j + 1 ]
= 2 2 m + 1 1 × [ ∑ i = 1 m ( 2 m + 1 m + i ) X 2 i + 1 + ∑ j = 1 m ( 2 m + 1 m + j ) Y 2 j + 1 ]
After substituting values for X i and Y j and using binomial theorem, we get:
Q = ( a 0 + α − 1 β ) 2 2 m + 1 ( α + α 1 ) 2 m + 1 − α − 1 β
Now putting α = φ , β = φ 1 , a 0 = 3 1 , 2 m + 1 = 5 1 , we get:
Q = 2 4 6 5 2 5 5 − 1
Thus, the required answer is 5 + 2 5 + 5 + 2 + 4 6 + 1 = 8 4