FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups 
 ProfileProfile   PreferencesPreferences   Log in to check your private messagesLog in to check your private messages   Log inLog in 
Forum index » Science and Technology » Math » Recreational
3 Dimensional Random Walk
Post new topic   Reply to topic Page 1 of 1 [6 Posts] View previous topic :: View next topic
Author Message
Chip Eastham
science forum Guru


Joined: 01 May 2005
Posts: 412

PostPosted: Tue May 30, 2006 3:29 pm    Post subject: Re: 3 Dimensional Random Walk Reply with quote

Jim Dars wrote:
Quote:
"Barry Schwarz" <schwarzb@doezl.net> wrote in message
news:ssoh72tir6bict6kvvqmhgqgpldp7mqo6r@4ax.com...
On Sat, 27 May 2006 13:32:50 -0700, "Jim Dars" <jim-dars@comcast.net
wrote:

Hi All,

Feller gives in Volume I, "An Introduction to Probability Theory and It's
Applications", that if one starts at the origin and moves at random the
probability of returning to the origin in one or two dimensions is one
and
in three dimensions it is about .35. (In two dimensions there are 4
random
choices and in 3 dimensions, 6.)

Question: In 3 dimensions if you start at the point (10000,10000,10000)
do
you arrive at a different answer?

Does your question refer to returning to the starting point or
returning to the origin?


Remove del for email

Hi Barry,

It concerns "returning" (getting there the first time) to the origin.

Actually, I suspect that, in the three dimensional case, the starting point
influences the probability of "returning" to the origin. I'm wondering what
the Group thinks (knows).

Best wishes, Jim

I can recommend this (free) monograph by Doyle and Snell on the
subject:

http://front.math.ucdavis.edu/math.PR/0001057

In two dimensions, a random walk on the "graph" connecting
the usual integer lattice points visits every point in the plane
with probability 1. So in that setting it doesn't matter whether
we ask about "returning" to the origin or going from one point
to another, the probability is the same. This can be deduced
from the divergence of SUM 1/n.

In three dimensions, as Jim quotes Feller, the chance of
returning to an original starting point is positive and strictly
less than 1 (about .35). This is related to the convergence
of SUM 1/(n^2), although the connection is not elegant when
made rigorous. The search for a simple connection motivates
much of the monograph above.

In short we can define a function that gives the probability
that a random walk starting at the origin will visit (at some
subsequent step) any specific point in the integer lattice.
Evaluation of this function at (1000,1000,1000) will give by
symmetry the answer to Jim's question, and in general
this function is monotone decreasing as we go farther
from the origin.

regards, chip
Back to top
Jim Dars
science forum beginner


Joined: 03 Jul 2005
Posts: 41

PostPosted: Wed May 31, 2006 9:03 pm    Post subject: Re: 3 Dimensional Random Walk Reply with quote

"Chip Eastham" <hardmath@gmail.com> wrote in message
news:1149002994.240544.237280@38g2000cwa.googlegroups.com...
Quote:

Jim Dars wrote:
"Barry Schwarz" <schwarzb@doezl.net> wrote in message
news:ssoh72tir6bict6kvvqmhgqgpldp7mqo6r@4ax.com...
On Sat, 27 May 2006 13:32:50 -0700, "Jim Dars" <jim-dars@comcast.net
wrote:

Hi All,

Feller gives in Volume I, "An Introduction to Probability Theory and
It's
Applications", that if one starts at the origin and moves at random
the
probability of returning to the origin in one or two dimensions is
one
and
in three dimensions it is about .35. (In two dimensions there are 4
random
choices and in 3 dimensions, 6.)

Question: In 3 dimensions if you start at the point
(10000,10000,10000)
do
you arrive at a different answer?

Does your question refer to returning to the starting point or
returning to the origin?


Remove del for email

Hi Barry,

It concerns "returning" (getting there the first time) to the origin.

Actually, I suspect that, in the three dimensional case, the starting
point
influences the probability of "returning" to the origin. I'm wondering
what
the Group thinks (knows).

Best wishes, Jim

I can recommend this (free) monograph by Doyle and Snell on the
subject:

http://front.math.ucdavis.edu/math.PR/0001057

In two dimensions, a random walk on the "graph" connecting
the usual integer lattice points visits every point in the plane
with probability 1. So in that setting it doesn't matter whether
we ask about "returning" to the origin or going from one point
to another, the probability is the same. This can be deduced
from the divergence of SUM 1/n.

In three dimensions, as Jim quotes Feller, the chance of
returning to an original starting point is positive and strictly
less than 1 (about .35). This is related to the convergence
of SUM 1/(n^2), although the connection is not elegant when
made rigorous. The search for a simple connection motivates
much of the monograph above.

In short we can define a function that gives the probability
that a random walk starting at the origin will visit (at some
subsequent step) any specific point in the integer lattice.
Evaluation of this function at (1000,1000,1000) will give by
symmetry the answer to Jim's question, and in general
this function is monotone decreasing as we go farther
from the origin.

regards, chip

Hi Chip,


Thank you. I suspected as much in that if you start at the origin you have
a 1/6 probability of reaching the origin on the second move. It's
interesting to learn that, as you move the starting point away from the
origin, the probability of reaching the origin diminishes. A little
reflection does indicate this as a logical conclusion, but seemingly logical
conclusions often blow up in your face.

Best wishes, Jim
Back to top
Mike Amling
science forum Guru


Joined: 05 May 2005
Posts: 525

PostPosted: Thu Jun 08, 2006 10:50 pm    Post subject: Re: 3 Dimensional Random Walk Reply with quote

In article <1149002994.240544.237280@38g2000cwa.googlegroups.com>,
Chip Eastham <hardmath@gmail.com> wrote:
Quote:

In short we can define a function that gives the probability
that a random walk starting at the origin will visit (at some
subsequent step) any specific point in the integer lattice.
Evaluation of this function at (1000,1000,1000) will give by
symmetry the answer to Jim's question, and in general
this function is monotone decreasing as we go farther
from the origin.

THe probability that a random walk starting at the origin will
eventually return to the origin should be the same as the probability
that a random walk starting at (1,0,0), (0,1,0), (0,0,1), (-1,0,0),
(0,-1,0), or (0,0,-1) will eventually reach the origin.

--
Please reply to: | "Any sufficiently advanced incompetence is
pciszek at panix dot com | indistinguishable from malice."
Autoreply is disabled |
Back to top
Google

Back to top
Display posts from previous:   
Post new topic   Reply to topic Page 1 of 1 [6 Posts] View previous topic :: View next topic
The time now is Sat Jun 24, 2017 8:43 am | All times are GMT
Forum index » Science and Technology » Math » Recreational
Jump to:  

Similar Topics
Topic Author Forum Replies Last Post
No new posts reference books for formulaes in stochastic analysis and ... Michael11 Math 0 Thu Jul 20, 2006 12:38 am
No new posts spectrum of a symmetric tridiagonal random matrix pf.buonsante@gmail.com Math 0 Wed Jul 19, 2006 9:45 am
No new posts random walk on graph Eric B num-analysis 1 Thu Jul 13, 2006 7:23 pm
No new posts Period of random number generators DGoncz@aol.com Math 9 Thu Jul 13, 2006 1:48 am
No new posts Order Statistic in Random Process/Random Sequence fan_bai@yahoo.com Math 0 Mon Jul 10, 2006 5:03 pm

Copyright © 2004-2005 DeniX Solutions SRL
Other DeniX Solutions sites: Electronics forum |  Medicine forum |  Unix/Linux blog |  Unix/Linux documentation |  Unix/Linux forums  |  send newsletters
 


Powered by phpBB © 2001, 2005 phpBB Group
[ Time: 0.0195s ][ Queries: 20 (0.0039s) ][ GZIP on - Debug on ]