Z statistic: Difference between revisions
imported>Doug Williamson (Align with qualifications material.) |
imported>Doug Williamson (Link with t-statistic page.) |
||
| Line 10: | Line 10: | ||
The Z score is calculated as the difference between the data point (X) and the mean E[x], all divided by the standard deviation of the population (SD). | The Z score is calculated as the difference between the data point (X) and the mean E[x], all divided by the standard deviation of the population (SD). | ||
For example if: | For example if: | ||
| Line 24: | Line 26: | ||
= - 1.64 standard deviations. | = - 1.64 standard deviations. | ||
In this case the Z score is negative, indicating that the data point (83.6) lies below the mean (of 100). | In this case the Z score is negative, indicating that the data point (83.6) lies below the mean (of 100). | ||
| Line 29: | Line 32: | ||
== See also == | == See also == | ||
* [[Standardised normal distribution]] | * [[Standardised normal distribution]] | ||
* [[t-statistic]] | |||
[[Category:Risk_frameworks]] | [[Category:Risk_frameworks]] | ||
Revision as of 06:57, 20 April 2015
A commonly used transformation of a normal distribution. The resulting standardised normal distribution has a mean of 0 and a standard deviation of 1.
It is used extensively in hypothesis testing.
Also known as the Z score.
So for example if a data point has a Z score (or Z statistic) of -1.64, then it lies 1.64 standard deviations below the mean.
The Z score is calculated as the difference between the data point (X) and the mean E[x], all divided by the standard deviation of the population (SD).
For example if:
the mean (E[x]) of a population = 100;
the standard deviation (SD) = 10; and
a given observation (or data point) = 83.6;
then the Z score (Z) is calculated as: Z = (X - E[x])/SD
= (83.6 - 100 = -16.4)/10
= - 1.64 standard deviations.
In this case the Z score is negative, indicating that the data point (83.6) lies below the mean (of 100).
