The mean may be conceived of as an estimate of the median. When the mean is not an accurate estimate of the median, the set of numbers, or frequency distribution, is said to be skewed.
We denote the set of data by X = {x1, x2, ..., xn}. The symbol µ (Greek: mu) is used to denote the arithmetic mean of a population. We use the name of the variable, X, with a horizontal bar over it as the symbol ("X bar") for a sample mean. Both are computed in the same way:
In certain situations, the arithmetic mean is the wrong concept of "average" altogether. For example, if a stock rose 10% in the first year, 30% in the second year and fell 10% in the third year, then it would be incorrect to report its "average" increase per year over this three year period as the arithmetic mean (10% + 30% + (-10%))/3 = 10%; the correct average in this case is the geometric mean which yields an average increase per year of only 8.8%.
If X is a random variable, then the expected value of X can be seen as the long-term arithmetic mean that occurs on repeated measurements of X. This is the content of the law of large numbers. As a result, the sample mean is used to estimate unknown expected values.
Note that several other "means" have been defined, including the generalized mean, the harmonic mean, the arithmetic-geometric mean, and the weighted mean.
See also: mean, average, summary statistics, variance, central tendency
(When used as a noun, the word "arithmetic" is pronounced with the accent on the second syllable, but when used in the present sense, as an adjective, the accent is on the third syllable: "arithMETic")