Re: Neumaier's Modification of Heisenberg 4: The "Generalized HUP"



>>From Osher Doctorow

Wait a moment! Isn't a random variable with a finite number of values
probabilistic?

Yes, but it's like throwing away knowledge/information when you replace
Individual values with means or "expectations". If all that remains of
probability-statistics in the quantum theory is finite discrete random
variables, then in addition to the incredible loss of knowledge from
aggregating or averaging Individual values, there is an incredible loss
of knowledge from throwing away all continuous random variables.

But aren't continuous random variables outlawed by discrete energy
eigenvalues for example? Heck no!

But it gets funnier to a mathematical probability-statistics person.
Many physicists and engineers don't know that there is a considerable
parallelism between continuous and discrete probability distributions!
Surely I'm joking, Mr. Feynman or no Mr. Feynman as the saying goes?
No. Of course, it helps to take a third year college probability
course at a research university or equivalent that requires calculus.

Look at these parallels:

1) uniform distribution and equiprobable (discrete) distribution

2. exponential distribution and geometric (discrete) distribution

3. gamma distribution and Poisson (discrete) distribution except that x
goes over to parameter a + 1 or alpha + 1

This is two out of three of the maximum entropy Shannon distributions
for continuous random variables! And the normal/Gaussian approximation
holds for most known discrete distributions, the normal/Gaussian being
the third Shannon maximum entropy distribution!

What this means in simple language is that not only Neumaier but nobody
can drop continuous probability from quantum theory without dropping
the whole quantum theory.

Next question :>)

Osher Doctorow

.



Relevant Pages

  • Re: Analysis Tool Pack: Random function help
    ... In Excel Help. ... Click the distribution method you want to use to create random values. ... Characterized by a probability of success on a given trial. ... Bernoulli random variables have the value 0 or 1. ...
    (microsoft.public.excel.programming)
  • Re: Good enough for crypto?
    ... >> random variables first as being deterministic, ... i.e. a random variable whose distribution ... can't be a constant (with probability 1), ... They're ingredients in the usual type of random process model ...
    (sci.crypt)
  • Re: Overlapping probability distributions
    ... What is the probability that a random man is taller ... of the two draws) the pair has the bivariate distribution A*B ... now reduces to a 1-dimensional integration. ... about linear combinations of independent normal random variables; ...
    (sci.math)
  • Re: Skewed differences and paired students t test
    ... > some circumstances) be both discrete and in a small range. ... The null distribution, evaluated this way, would be based on ... careful to calculate the right probability for the significance level: ... pairs is small then you could replace the randomisation by a complete ...
    (sci.stat.math)
  • Re: continuous or discrete?
    ... The random variable is neither continous nor discrete. ... consider a customer's waiting time when s/he enters a store. ... For an example, consider the Cantor function as a cumulative distribution function on. ... Such random variables rarely come up in modelling. ...
    (sci.math)