Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

What is the significance of cumulative normal = 1?

I am trying to program a game, however I have inherited code from another programmer. He is using the cumulative norm. his code is outputting a value of one for a certain value x, mean and standard deviation.

I have a product value which lets say is = 2.5. The standard deviation is 0.625. The x value was generated iteratively and eventually comes to 10.931456.

So are you saying that X or the mean is the 2.5, x is the 10.93, and the sd = 0.625?

Also that the probability of 2.5 being at the point x = 10.93 is 1?

share|improve this question

1 Answer 1

up vote 0 down vote accepted

If I understand your question correctly, you have a normally distributed random variable $X$, and the program says that the probability that $X\le x$ is $1$. Now for a normally distributed random variable $X$, the probability that $X\le x$ can never be exactly equal to $1$, it must always be less than $1$.

However, the tails decay very rapidly, so we may have $P(X\le x)$ for all practical purposes. So for example we could easily have $P(X\le x)=1-10^{-12}$. For instance, the probability that a standard normal $Z$ is $\le 6$ is about $1-2\times 10^{-9}$.

The programmer may have decided to round such a number to $1$. Or else, because of precision limitations, there is automatic rounding.

Deciding to round is very sensible. When you are modelling a real phenomenon using the normal distribution, that distribution will never model the phenomenon exactly. Sort of close is the best one can hope for.

Added: For your edited question, I do not know what the $2.5$ means. But the range from $2.5$ to $10.93$ is $8.43$, which is $8.43/0.625$ standard deviation units, about $13.5$ standard deviation units. That is an astoundingly large range, really quite inconsistent with a normal distribution. For most practical (or even impractical) purposes, in a normal there is virtually nothing $4$ standard deviation units or more away from the mean.

share|improve this answer
    
Yes it is assumed one since the value gets closer and closer to 1. And yes because of precision limits. Can you see my edit above? –  user1064028 Feb 4 '12 at 17:18
    
So is my edit above (in the original question) correct? –  user1064028 Feb 4 '12 at 17:34
    
@user1064028: I added some stuff to the answer. I cannot know what the numbers stand for, except for the standard deviation. It makes sense that the $x$ is $10.93$. The number $2.5$ cannot be "at" $10.93$, that doesn't make sense. The random variable $X$ (the numerical result of the experiment) can in principle be $\ge 10.93$, though with a normal with SD $0.625$, and mean $2.5$, or even a mean as large as $8$, this can basically never happen. That is consistent with the computed cumulative distribution function being virtually $1$ at $10.93$. –  André Nicolas Feb 4 '12 at 17:51
    
Ok thank you it is making sense now –  user1064028 Feb 4 '12 at 17:58

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.