Average

In mathematics, an average, mean, or central tendency of a data set refers to a measure of the
"middle" or "expected" value of the data set. There are many different descriptive statistics that
can be chosen as a measurement of the central tendency. The most common method, and the one
generally referred to simply as the average, is the arithmetic mean. Please see the table of
mathematical symbols for explanations of the symbols used.
In statistics, the term central tendency is used in some fields of empirical research to refer to
what statisticians sometimes call "location". A "measure of central tendency" is either a location
parameter or a statistic used to estimate a location parameter.
Measures of central tendency
There are several different kinds of calculations for central tendency, the kind of calculation that
should be used depends on the type of data level of measurement! and purpose for which the
central tendency is being calculated"
Name Equation or description
Arithmetic
mean
#edian The middle value that separates the higher half from the lower half of the data set
#ode The most fre$uent value in the data set
%eometric
mean
&armonic
mean
'uadratic
mean
or (#)!
%enerali*ed
mean
&eronian
mean
+eighted
mean
Truncated
mean
The arithmetic mean of data values after a certain number or proportion of the
highest and lowest data values have been discarded
Inter$uartile
mean
A special case of the truncated mean, using the inter$uartile range
#idrange
+insori*ed
mean
)imilar to the truncated mean, but, rather than deleting the extreme values, they
are set e$ual to the largest and smallest values that remain
Other averages
,ther more sophisticated averages are" trimean, trimedian, and normali*ed mean. These are
usually more representative of the whole data set.
-citation needed.
,ne can create one/s own average metric using %eneralised f0mean"
where f is any invertible function. The
harmonic mean is an example of this using fx! 1 2 3 x, and the geometric mean is another, using
fx! 1 logx. Another example, expmean exponential mean! is a mean using the function fx! 1 e
x
,
and it is inherently biased towards the higher values.
The only significant reason why the arithmetic mean classical average! is generally used in
scientific papers is that there are various statistical! tests which can be applied to test the
statistical significance of the results, as well as the correlations that are explored through these
metrics.
-citation needed.
Average applied to a Data Stream
The concept of an average can be applied to a stream of data as well as a bounded set, the goal
being to find a value about which recent data is in some way clustered. The stream may be
distributed in time, as in samples taken by some data ac$uisition system from which we want to
remove noise, or in space, as in pixels in an image from which we want to extract some property.
An easy0to0understand and widely used application of average to a stream is the simple moving
average in which we compute the arithmetic mean of the most recent 4 data items in the stream.
To advance one position in the stream, we add 234 times the new data item and subtract 234
times the data item 4 places back in the stream.
Derivation of the name
The original meaning of the word is "damage sustained at sea"" the same word is found in
Arabic as awar, in Italian as avaria and in 5rench as avarie. &ence an average adjuster is a
person who assesses an insurable loss.
#arine damage is either particular average, which is borne only by the owner of the damaged
property, or general average, where the owner can claim a proportional contribution from all the
parties to the marine venture. The type of calculations used in ad6usting general average gave
rise to the use of "average" to mean "arithmetic mean".
Statistical dispersion
In mathematics, a (statistical) dispersion also called statistical variability! of a set list! of
data is a measure how observations in the data set are distributed across various categories.
There are many different descriptive statistics that can be chosen as a measurement of the central
tendency. In other words, dispersion is $uantifiable variation of measurements of differing
members of a population within the scale on which they are measured.
Measures of statistical dispersion
A measure of statistical dispersion is a real number that is *ero if all the data are identical, and
increases as the data becomes more diverse. An important measure of dispersion is the standard
deviation, which is the s$uare root of the variance which is itself a measure of dispersion!.,ther
such measures include the range, the inter$uartile range, the mean difference, and the average
absolute deviation, and, in the case of categorical random variables, the discrete entropy. 4one
of these can be negative7 their least possible value is *ero.
A measure of statistical dispersion is particularly useful if it is location0invariant, and linear in
scale. )o if a random variable X has a dispersion of S
X
then a linear transformation Y 1 aX 8 b
for real a and b should have dispersion S
Y
1 9a9S
X
. ,ne of the forms in which statistical
variability is reali*ed in the empirical sciences is that of differences in repeated measurements of
the same $uantity.
Sources of statistical dispersion
In the physical sciences, such variability may result only from random measurement errors"
instrument measurements are often not perfectly precise 0 i.e., reproducible. ,ne may assume
that the $uantity being measured is unchanging and stable, and that the variation between
measurements is due to observational error.
In the biological sciences, this assumption is false" the variation observed might be intrinsic to
the phenomenon" distinct members of a population differ greatly. This is also seen in the arena of
manufactured products7 even there, the meticulous scientist finds variation.
The simple model of a stable $uantity is preferred when it is tenable. :ach phenomenon must be
examined to see if it warrants such a simplification.
Summary statistics
In descriptive statistics, summary statistics are used to summari*e a set of observations, in
order to communicate as much as possible as simply as possible. )tatisticians commonly try to
describe the observations in
2. a measure of location, or central tendency, such as the arithmetic mean, median, mode, or
inter$uartile mean
;. a measure of statistical dispersion like the standard deviation, variance, range, or
inter$uartile range, or absolute deviation.
<. a measure of the shape of the distribution like skewness or kurtosis
The %ini coefficient was originally developed to measure income ine$uality, but can be used for
other purposes as well.
Standard deviation
In probability and statistics, the standard deviation of a probability distribution, random
variable, or population or multiset of values is a measure of the spread of its values. It is defined
as the s$uare root of the variance.
The standard deviation is the root mean s$uare (#)! deviation of values from their arithmetic
mean. 5or example, in the population =>, ?@, the mean is A and the standard deviation is ;. This
may be written" =>, ?@ B AC;. In this case 2DDE of the values in the population are at one
standard deviation of the mean.
The standard deviation is the most common measure of statistical dispersion, measuring how
widely spread the values in a data set are. If the data points are close to the mean, then the
standard deviation is small. Fonversely, if many data points are far from the mean, then the
standard deviation is large. If all the data values are e$ual, then the standard deviation is *ero.
The standard deviation G! of a population can be estimated by a modified standard deviation s!
of a sample. The formulae are given below.
Definition and calculation
Standard deviation of a random variable
The standard deviation of a random variable X is defined as"
where :X! is the expected value of X.
4ot all random variables have a standard deviation, since these expected values need not exist.
5or example, the standard deviation of a random variable which follows a Fauchy distribution is
undefined.
If the random variable X takes on the values x
2
,...,x
N
which are real numbers! with e$ual
probability, then its standard deviation can be computed as follows. 5irst, the mean of X, , is
defined as"
see sigma notation! where 4 is the number of samples taken. 4ext, the standard deviation
simplifies to"
In other words, the standard deviation of a discrete uniform random variable X can be calculated
as follows"
2. 5or each value x
i
calculate the difference between x
i
and the average value .
;. Falculate the s$uares of these differences.
<. 5ind the average of the s$uared differences. This $uantity is the variance G
;
.
>. Take the s$uare root of the variance.
Estimating population standard deviation from sample standard deviation
In the real world, finding the standard deviation of an entire population is unrealistic except in
certain cases, such as standardi*ed testing, where every member of a population is sampled. In
most cases, the standard deviation is estimated by examining a random sample taken from the
population. The most common measure used is the sample standard deviation, which is defined
by
where is the sample and is the mean of the sample.
The reason for this definition is that s
;
is an unbiased estimator for the variance G
;
of the
underlying population, if that variance exists and the sample values are drawn independently
with replacement. &owever, s is not an unbiased estimator for the standard deviation G7 it tends
to underestimate the population standard deviation. Although an unbiased estimator for G is
known when the random variable is normally distributed, the formula is complicated and
amounts to a minor correction. #oreover, unbiasedness, in this sense of the word, is not always
desirable7 see bias of an estimator.
Another estimator sometimes used is the similar expression
This form has a uniformly smaller mean s$uared error than does the unbiased estimator, and is
the maximum0likelihood estimate when the population is normally distributed.
Eample
+e will show how to calculate the standard deviation of a population. ,ur example will use the
ages of four young children" = H, A, ?, I @.
)tep 2. Falculate the mean average, "
+e have N 1 > because there are four data points"
(eplacing N with >
This is the mean.
)tep ;. Falculate the standard deviation "
(eplacing N with >
(eplacing with J
)o, the standard deviation is the s$uare root of five halves, or approximately 2.H?22.
+ere this set a sample drawn from a larger population of children, and the $uestion at hand was
the standard deviation of the population, convention would replace the N or >! here with NK2
or <!.
!nterpretation and application
A large standard deviation indicates that the data points are far from the mean and a small
standard deviation indicates that they are clustered closely around the mean.
5or example, each of the three data sets D, D, 2>, 2>!, D, A, ?, 2>! and A, A, ?, ?! has a mean of
J. Their standard deviations are J, H, and 2, respectively. The third set has a much smaller
standard deviation than the other two because its values are all close to J. In a loose sense, the
standard deviation tells us how far from the mean the data points tend to be. It will have the
same units as the data points themselves. If, for instance, the data set D, A, ?, 2>! represents the
ages of four siblings, the standard deviation is H years.
As another example, the data set 2DDD, 2DDA, 2DD?, 2D2>! may represent the distances traveled
by four athletes in < minutes, measured in meters. It has a mean of 2DDJ meters, and a standard
deviation of H meters.
In the age example above, a standard deviation of H may be considered large7 in the distance
example above, H may be considered small small to the mathematician, not so small to the
athletes!.
)tandard deviation may serve as a measure of uncertainty. In physical science for example, the
reported standard deviation of a group of repeated measurements should give the precision of
those measurements. +hen deciding whether measurements agree with a theoretical prediction,
the standard deviation of those measurements is of crucial importance" if the mean of the
measurements is too far away from the prediction with the distance measured in standard
deviations!, then we consider the measurements as contradicting the prediction. This makes
sense since they fall outside the range of values that could reasonably be expected to occur if the
prediction were correct and the standard deviation appropriately $uantified. )ee prediction
interval.
"eal#life eamples
The practical value of understanding the standard deviation of a set of values is in appreciating
how much variation there is away from the "average" mean!.
Weather
As a simple example, consider average temperatures for cities. +hile the average for all cities
may be ADL5, it/s helpful to understand that the range for cities near the coast is smaller than for
cities inland, which clarifies that, while the average is similar, the chance for variation is greater
inland than near the coast.
)o, an average of AD occurs for one city with highs of ?DL5 and lows of >DL5, and also occurs for
another city with highs of AH and lows of HH. The standard deviation allows us to recogni*e that
the average for city with the wider variation, and thus a higher standard deviation will not offer
as reliable a prediction of temperature as the city with the smaller variation and lower standard
deviation.
Sports
Another way of seeing it is to consider sports teams. In any set of categories, there will be teams
that rate highly at some things and poorly at others. Fhances are, the teams that lead in the
standings will not show such disparity, but will be pretty good in most categories. The lower the
standard deviation of their ratings in each category, the more balanced and consistent they might
be. )o, a team that is consistently bad in most categories will have a low standard deviation
indicating they will probably lose more often than win. A team that is consistently good in most
categories will also have a low standard deviation and will therefore end up winning more than
losing. A team with a high standard deviation might be the type of team that scores a lot strong
offense! but gets scored on a lot too weak defense!7 or vice versa, might get scored on, but
compensate with higher scoring 0 teams with a higher standard deviation will be more
unpredictable.
Trying to predict which teams, on any given day, will win, may include looking at the standard
deviations of the various team "stats" ratings, in which anomalies can match strengths vs
weaknesses to attempt to understand what factors may prevail as stronger indicators of eventual
scoring outcomes.
In racing, a driver is timed on successive laps. A driver with a low standard deviation of lap
times is more consistent than a driver with a higher standard deviation. This information can be
used to help understand where opportunities might be found to reduce lap times.
Finance
In finance, standard deviation is a representation of the risk associated with a given security
stocks, bonds, etc.!, or the risk of a portfolio of securities. (isk is an important factor in
determining how to efficiently manage a portfolio of investments because it determines the
variation in returns on the asset and3or portfolio and gives investors a mathematical basis for
investment decisions. The overall concept of risk is that as it increases, the expected return on
the asset will increase as a result of the risk premium earned 0 in other words, investors should
expect a higher return on an investment when said investment carries a higher level of risk.
5or example, you have a choice between two stocks" )tock A historically returns HE to investors
with a standard deviation of 2DE, )tock M historically returns AE to investors and carries a
standard deviation of ;DE. ,n the basis of risk and return, )tock A is the acceptable choice
because earning an extra 2E with )tock M is not worth double the amount of risk as )tock A. In
other words, )tock M is likely to lose money for the investor more often than )tock A will under
the same circumstances, and will only return 2E more than )tock A. In this example, )tock A
has the potential to earn 2DE more than the expected return, but is e$ually as likely to lose 2DE
less than the expected return.
5or a set of returns of a security, calculating the average return arithmetic mean! of the security
over a given number of periods will derive the expected return on the asset. 5or each individual
return period, subtracting the expected return from the actual return in the period will result in
the variance, or the difference between what you expected to earn and what you actually earned.
)$uare the variance in each period to find the effect that the result has on the overall risk of the
asset 0 the larger the variance in a period the greater risk that security carries. Taking the average
of the s$uared variances results in the measurement of overall units of risk associated with the
asset. 5inding the s$uare root of this variance will result in the standard deviation of the
investment tool in $uestion. Nse this measurement, combined with the average return on the
security, as a basis for analysis when comparing two or more securities.
$eometric interpretation
To gain some geometric insights, we will start with a population of three values, x
2
, x
;
, x
<
. This
defines a point P 1 x
2
, x
;
, x
<
! in "
<
. Fonsider the line L 1 =r, r, r! " r in "@. This is the "main
diagonal" going through the origin. If our three given values were all e$ual, then the standard
deviation would be *ero and P would lie on L. )o it is not unreasonable to assume that the
standard deviation is related to the distance of P to L. And that is indeed the case. #oving
orthogonally from P to the line L, one hits the point"
whose coordinates are the mean of the values we started out with. A little algebra shows that the
distance between P and R which is the same as the distance between P and the line L! is given
by GO. An analogous formula with < replaced by N! is also valid for a population of N values7
we then have to work in "
N
.
"ules for normally distributed data
Park blue is less than one standard deviation from the mean. 5or the normal distribution, this
accounts for A?.;JE of the set7 while two standard deviations from the mean medium and dark
blue! account for IH.>HE7 and three standard deviations light, medium, and dark blue! account
for II.J<E.
In practice, one often assumes that the data are from an approximately normally distributed
population. If that assumption is 6ustified, then about A?E of the values are within 2 standard
deviation of the mean, about IHE of the values are within two standard deviations and about
II.JE lie within < standard deviations. This is known as the %&'#()#((*+ rule%, or %the
empirical rule%
The confidence intervals are as follows"
G A?.;A?I>I;2<J2E
;G IH.>>IIJ<A2D<AE
<G II.J<DD;D<I<AJE
>G II.II<AAHJH2A<E
HG II.IIII>;AAIAIE
AG II.IIIIII?D;A?E
JG II.IIIIIIIIIJ>E
5or normal distributions, the two points of the curve which are one standard deviation from the
mean are also the inflection points.
,hebyshev rules
If it is not known whether the distribution is normal, one can always use Fhebyshev/s ine$uality"
At least HDE of the values are within 2.> standard deviations from the mean.
At least JHE of the values are within ; standard deviations from the mean.
At least ?IE of the values are within < standard deviations from the mean.
At least I>E of the values are within > standard deviations from the mean.
At least IAE of the values are within H standard deviations from the mean.
At least IJE of the values are within A standard deviations from the mean.
At least I?E of the values are within J standard deviations from the mean.
At least 2DD Q 2 0 23!
;
! percent of the values are within ! standard deviations from the
mean.
"elationship bet-een standard deviation and mean
The mean and the standard deviation of a set of data are usually reported together. In a certain
sense, the standard deviation is a "natural" measure of statistical dispersion if the center of the
data is measured about the mean. This is because the standard deviation from the mean is
smaller than from any other point. The precise statement is the following" suppose x
2
, ..., x
n
are
real numbers and define the function"
Nsing calculus, it is possible to show that Gr! has a uni$ue minimum at the mean"
this can also be done with fairly simple algebra alone, since, as a function of r, it is a $uadratic
polynomial!.
The coefficient of variation of a sample is the ratio of the standard deviation to the mean. It is a
dimensionless number that can be used to compare the amount of variance between populations
with different means.
Fhebyshev/s ine$uality proves that in any data set, nearly all of the values will be nearer to the
mean value, where the meaning of "close to" is specified by the standard deviation.
"apid calculation methods
A slightly faster significantly for running standard deviation! way to compute the sample
standard deviation is given by the following formula though this can exacerbate round0off
error!"
)imilarly for population standard deviation"
,r from running sums"