Right up and away

Suppose a particle, starting at the origin, moves in six positive integral steps, (i.e., each step is of integral length 1 \ge 1 ), in the pattern right, up, right, up, right, up, such that the combined lengths of the steps is 10 10 , and such that each possible sequence of steps is equally likely to occur.

If D D is the expected (magnitude of the) distance between the origin and the particle after it has completed the six steps, find 1000 D . \lfloor 1000*D \rfloor.

Clarifications :

By "right" I mean in the positive x x -direction, and by "up" I mean in the positive y y -direction.

As an example, one possible path is 2 2 right, 2 2 up, 1 1 right, 3 3 up, 1 1 right and 1 1 up.


The answer is 7267.

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.

2 solutions

First, with r k r_{k} representing the length of the k k th right step and u j u_{j} representing the length of the j j th up step, we need to look at the solutions of the equation

r 1 + u 1 + r 2 + u 2 + r 3 + u 3 = 10. r_{1} + u_{1} + r_{2} + u_{2} + r_{3} + u_{3} = 10.

Now since each of these variables must be 1 , \ge 1, we can let each r k = r k 1 r_{k}' = r_{k} - 1 and u j = u j 1 , u_{j}' = u_{j} - 1, and instead look at the equation

r 1 + u 1 + r 2 + u 2 + r 3 + u 3 = 4 , r_{1}' + u_{1}' + r_{2}' + u_{2}' + r_{3}' + u_{3}' = 4,

where each of the variables is non-negative. This is a stars and bars equation with solution ( 9 4 ) = 126. \dbinom{9}{4} = 126.

Now what matters here as far as calculating D D is concerned is the total distances right and up the particle moves after six steps. So we can have k = 1 3 r k = 0 , 1 , 2 , 3 , 4 \sum_{k=1}^{3} r_{k}' = 0,1,2,3,4 with corresponding values j = 0 4 u j = 4 , 3 , 2 , 1 , 0. \sum_{j=0}^{4} u_{j}' = 4,3,2,1,0. So now we have a series of additional stars and bars equations to solve. For example, with the sum of the r k r_{k}' s being 3 3 , we are solving the equation r 1 + r 2 + r 3 = 3 , r_{1}' + r_{2}' + r_{3}' = 3, which has the solution ( 5 3 ) = 10. \binom{5}{3} = 10. The corresponding u j u_{j} equation is u 1 + u 2 + u 3 = 1 , u_{1}' + u_{2}' + u_{3}' = 1, which has solution ( 3 1 ) = 3. \binom{3}{1} = 3. Thus there are 10 3 = 30 10*3 = 30 paths that, after the six steps are completed, involve the particle going 6 6 units right and 4 4 units up, which corresponds to a distance of 6 2 + 4 2 = 2 13 . \sqrt{6^{2} + 4^{2}} = 2\sqrt{13}.

Similarly, there are:

  • 15 15 paths each where the particle either moves 7 7 units to the right and 3 3 units up, or 3 3 units to the right and 7 7 units up, resulting in a distance of 49 + 9 = 58 \sqrt{49 + 9} = \sqrt{58} ;

  • 30 30 paths each where the particle either moves 6 6 units right and 4 4 units up, or 4 4 units right and 6 6 units up, resulting in a distance of 2 13 2\sqrt{13} ;

  • 36 36 paths where the particle goes both 5 5 units right and 5 5 units up, for a distance of 5 2 . 5\sqrt{2}.

Thus D = 30 126 58 + 60 126 ( 2 13 ) + 36 126 ( 5 2 ) = D = \dfrac{30}{126}\sqrt{58} + \dfrac{60}{126}(2\sqrt{13}) + \dfrac{36}{126}(5\sqrt{2}) =

5 21 58 + 20 21 13 + 10 7 2 = 7.26744.... \dfrac{5}{21}\sqrt{58} + \dfrac{20}{21}\sqrt{13} + \dfrac{10}{7}\sqrt{2} = 7.26744....

Thus 1000 D = 7267 . \lfloor 1000*D \rfloor = \boxed{7267}.

I did it in a similar manner, partitioned 10 into two, and then each of them further into three.

Raghav Vaidyanathan - 6 years, 1 month ago

Log in to reply

Great. I figured that there would be a few ways to do this, but that they all involved a fair bit of case work. I'm glad that you and others reached the same answer as I did; I wasn't 100% sure when I posted the question.

Brian Charlesworth - 6 years, 1 month ago

Log in to reply

Actually my method doesn't involve casework. It is exactly same as yours, but with the exception that I applied "stars and bars" on natural numbers rather than whole numbers. I don't understand why it is taught that way.

Raghav Vaidyanathan - 6 years, 1 month ago
Trevor Arashiro
Apr 30, 2015

Thanks to Brian for helping me fix this.

By stars and bars, we have the number of ways to go 3-7 units right and 7-3 units up is

2 6 ( n 2 ) ( 8 n 2 ) = 126 \displaystyle \sum_2^6 \dbinom{n}{2}\dbinom{8-n}{2}=126

The total length traveled by each of these will be

( n + 1 ) 2 + ( 9 n ) 2 \sqrt{(n+1)^2+(9-n)^2}

So we evaluate the sum and divide by the total number of cases to get to be (note that we begin the sum at n=2 and not 3 so that's why the sqrt changes)

2 6 ( n 2 ) ( 8 n 2 ) ( n + 1 ) 2 + ( 9 n ) 2 2 6 ( n 2 ) ( 8 n 2 ) 7.397 \dfrac{\displaystyle \sum_2^6 \dbinom{n}{2}\dbinom{8-n}{2}\sqrt{(n+1)^2+(9-n)^2}}{\displaystyle \sum_2^6 \dbinom{n}{2}\dbinom{8-n}{2}}\approx 7.397

I think that the stars and bars part is right, but that the length function should be

( n + 1 ) 2 + ( 10 ( n + 1 ) ) 2 = ( n + 1 ) 2 + ( 9 n ) 2 . \sqrt{(n + 1)^{2} + (10 - (n + 1))^{2}} = \sqrt{(n + 1)^{2} + (9 - n)^{2}}.

I did a WA check and with this change we get the same answer. Basically we've done it the same way, but I used lots of words to explain how the method works and you cut to the chase and created a nice formula. :)

Brian Charlesworth - 6 years, 1 month ago

Log in to reply

Ahhhh, That makes sense, since the beginning should technically be ( n 1 2 ) \dbinom{n-1}{2} . But I changed the (what ever the numbers at the top and bottom of the sigma sign are called) thingys. Is there a name for those things?

Trevor Arashiro - 6 years, 1 month ago

Log in to reply

Haha. Those "thingys" are the starting, (or upper), and ending, (or lower), indices, (with n n being the index variable). :) I suppose you should also make the change to the expression for the length function the first time you mention it, just to be complete.

Brian Charlesworth - 6 years, 1 month ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...