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.

Along with the knowledge of the shape of a given probability distribution - what practical information does the mean and standard deviation of a random variable provide?

share|improve this question
add comment

3 Answers

up vote 1 down vote accepted

The mean is measure of location and the standard deviation is a measure of scale.

If a random variable $X$ has mean $\mu$ and standard deviation $\sigma$ and you use define a new random variable as $Y=X+a$ then the mean of $Y$ is $\mu+a$. Similarly if you define $Z = bX$ then the standard deviation of $Z$ is $b\sigma$. More generally for a linear transformation, if $W=bX+a$ then the mean of $W$ is $b\mu +a$ and the standard deviation of $W$ is $b\sigma$.

There are other statistics with similar properties, such as the median and inter-quartile range.

If the mean and standard deviation exist then the distribution is not "too dispersed", and you can apply the central limit theorem.

share|improve this answer
add comment
  • Mean tells you around which point is the distribution concentrated on.

  • Standard deviation tells you, hoe dispersed is the distribution around the mean. The greater the dispersion of values, the larger the standard deviation.

share|improve this answer
add comment

To flesh out Kannappan Sampath's answer a little, you can use Chebyshev's inequality to assert that the probability that $X$ deviates from its mean $\mu$ by at least $\alpha$ standard deviations is at most $1/\alpha^2$, that is,

$$P\{X \leq \mu - \alpha\sigma\} + P\{X \geq \mu + \alpha\sigma\} = P\{|X-\mu| \geq \alpha\sigma\} \leq \frac{1}{\alpha^2}$$ where $\sigma$ denotes the standard deviation.

share|improve this answer
add comment

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.