Talk:Rademacher distribution

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Mathematics (Rated Start-class, Low-importance)
WikiProject Mathematics
This article is within the scope of WikiProject Mathematics, a collaborative effort to improve the coverage of Mathematics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
Mathematics rating:
Start Class
Low Importance
 Field: Probability and statistics
WikiProject Statistics (Rated Start-class, Mid-importance)
WikiProject icon

This article is within the scope of the WikiProject Statistics, a collaborative effort to improve the coverage of statistics on Wikipedia. If you would like to participate, please visit the project page or join the discussion.

Start-Class article Start  This article has been rated as Start-Class on the quality scale.
 Mid  This article has been rated as Mid-importance on the importance scale.
 

Inconsistent use of mode in table[edit]

In Bernoulli distribution the definition as on http://en.wikipedia.org/wiki/Mode_(statistics) is used (all the values with max probability mass are shown). Here it says NA, whereas the situation is identical. If someone can check my thinking on this, it can be fixed (don't want to fix it if I am overlooking something simple). Kasterma (talk) 12:34, 18 August 2014 (UTC)

Unclear section on bounds[edit]

A section added today entitled Bounds on sums says

Let x be a random variable with a Rademacher distribution. Let yi be a sequence of real numbers. Then
  P( \sum ( x y_i ) > t || y ||_2 ) \le e^{ \frac{ t^2 }{ 2 } }


where || ||2 is the quadratic norm and P(a) is the probability of event a.

I have various problems with this. First, on Wikipedia we denote random variables with capital letters (X instead of x). Second, t needs to be defined and given a range. Third, the summation is unclear: is the random variable X taking on various values xi (in which case it should be xi in the summation) so that we are simply taking a weighted sum of independently drawn values of X? Or is it something else? This needs to be clarified. Fourth, this inequality says that a probability is less than or equal to something that is always greater than 1, which is always true of probabilities and so provides no information about this particular distribution. Fifth, y is not defined -- is it supposed to be the vector with elements yi, so that ||y||2 is the square root of the sum of the squares of the yi? Sixth, while the section title mentions bounds on sums, this is apparently intended to be a bound on a probability.

Then the new section says

If ||yi||1 is finite then
 P( \sum ( x y_i ) > t || y ||_1 ) = 0

The same questions apply to this, and also the norm ||.||1 needs to be explicitly defined.

Given that the new section is currently uninterpretable, I'm reverting it. I hope that it will be substantially clarified and reinserted. Duoduoduo (talk) 16:54, 14 May 2013 (UTC)