Question on Runs

From: Sriram (srinsriram_at_gmail.com)
Date: 10/27/04 

Date: 27 Oct 2004 11:04:52 -0700

Suppose I have 100 responses in an experiment, 50 left and 50 right.

I select, randomly without replacement, from these 100 responses and
assign to trials 1 to 100.

Barring the first trial, the other 99 trials either represent a
response shift (left to right or vice versa) from the prior trial or
represent an absence of response shift.

1. Will simple random sampling without replacement ensure that the
probability of a shift is around 50% ?

2. If the answer to 1 is NO, then what is the probability ? Is there
an algorithm that will easily generate a psuedo random sequence that
has a shift probability close to 50.

I assume this involves the hypergeometric distribution and would
appreciate pointers to any relevant literature.

thanks

sriram

srinsriram AT google mail

Relevant Pages

  • Re: So called "stimulus/response" models
    ... Instead of answering to each misunderstood, ironic and out of context ... Sorry, you exhibit a simplistic view of probability theory, and an even more ... of acquiring the consequences of responses. ... distribution over consequences of a given act. ...
    (comp.ai.philosophy)
  • Re: So called "stimulus/response" models
    ... claiming that all learning is operant conditioning. ... of acquiring the consequences of responses. ... distribution over consequences of a given act. ... Sorry, you exhibit a simplistic view of probability theory, and an even more ...
    (comp.ai.philosophy)
  • Re: The fun with the "categorization"
    ... This tends to make the words a sort of single stimulus class. ... hearing or reading these words alter the ... probability of responses that tend to occur with them in discussions, ...
    (sci.cognitive)
  • Re: Yet another Attempt at Disproving the Halting Problem
    ... > What is the probability that the ad verecundiam fallacy is not an error of reasoning? ... * Expert community may in fact agree that X is true. ... responses that have been given to you. ... computer scientists who responded to you have a poorer understanding ...
    (comp.theory)
  • Re: Yet another Attempt at Disproving the Halting Problem
    ... > What is the probability that the ad verecundiam fallacy is not an error of reasoning? ... * Expert community may in fact agree that X is true. ... responses that have been given to you. ... computer scientists who responded to you have a poorer understanding ...
    (comp.lang.cpp)