# 'normally distributed random numbers' vs 'uniformly distributed random number'?

what is the difference between 'normally distributed random numbers' and 'uniformly distributed random number'? A answer in a layman language is appreciated :)

• Uniformly distributed random numbers on an interval have equal probability of being selected or happening. Normally distributed random numbers on an interval have probabilities that follow the normal distribution bell curve, so numbers closer to the mean are more likely to be selected or to happen. Commented Jan 30, 2014 at 14:05

The green line shows a uniform distribution over the range $[-5, 5]$. Informally, each number in the range is equally ("uniformly") likely to be picked.

The red line shows a normal distribution with mean of 0 and standard deviation of 1. Numbers close to the mean are much more likely to be picked than those far away from the mean, in a particular and very special way.

The special thing about the normal distribution is this: If you take a large number of samples from a population with any distribution (subject to some not very strict conditions) and average them, the resulting distribution will approximate a normal distribution. For example, if you roll many dice and average the result, the resulting number will be distributed normally. The more dice you use, the closer the result will be to a normal distribution.

This property is why the normal distribution appears in nature. People's heights, for example, are normally distributed, because there are a large number of random factors that affect a person's height, but when they're all added together, the result is normal.

• what does Standard deviation mean (1) in this graph? Commented Jan 30, 2014 at 14:57
• The standard deviation measures how spread-out the distribution is. When the standard deviation is large, there are many values that are far from the mean; when it is small, most of the values cluster closely around the mean.
– MJD
Commented Jan 30, 2014 at 15:02
• @MJD: I don't understand when you say: "if you roll many dice and average the result, the resulting number will be distributed normally", but isn't a dice e uniform distribution rather than a normal one?! I mean every face of the dice is supposed to be equally probable. Commented Aug 18, 2015 at 13:21
• @marco If you roll even as few as 2 dice and average them, the result is not uniformly distributed. The result is twice as likely to be 3.5 as it is to be 2.
– MJD
Commented Aug 18, 2015 at 13:31
• @MJD: thanks now I have understood. Commented Aug 18, 2015 at 15:24