Calculate arithmetic mean, weighted average, geometric mean, harmonic mean, and moving average from any dataset. Shows all averages simultaneously with deviation analysis.
| # | Value | Deviation | |Deviation| |
|---|
The word "average" is one of the most commonly used β and most commonly misunderstood β terms in everyday mathematics. In casual conversation, "average" usually means the arithmetic mean (add up all values, divide by the count). But statisticians use at least five different types of average, each appropriate for different situations. This free average calculator computes all five simultaneously: arithmetic mean, weighted average, geometric mean, harmonic mean, and moving average.
The arithmetic mean is the sum of all values divided by the number of values:
Mean = (xβ + xβ + β¦ + xβ) / n
It is the most widely used average. The arithmetic mean of {85, 92, 78, 96, 88} is (85+92+78+96+88)/5 = 439/5 = 87.8.
The arithmetic mean is sensitive to outliers. If Bill Gates walks into a room of 10 people with average income $50,000, the arithmetic mean of the 11 people shoots up dramatically β even though 10 of the 11 are still earning $50,000. This is why income statistics in the US and UK typically report median income rather than mean income.
A weighted average gives different values different levels of importance (weight). The formula is:
Weighted Mean = Ξ£(xα΅’ Γ wα΅’) / Ξ£wα΅’
The most common real-world application is calculating a grade point average (GPA) or overall exam score. If your final exam counts for 50% of your grade and coursework for 50%, a weighted average gives the correct final score.
Example: Course grades with weights
In the UK, A-Level grades are often computed as weighted combinations of component papers. In the US, GPA calculations typically weight courses by their credit hours.
The geometric mean is the nth root of the product of n values:
Geometric Mean = (xβ Γ xβ Γ β¦ Γ xβ)^(1/n)
The geometric mean is ideal for averaging rates, ratios, and growth factors. It is the correct average for investment returns over multiple periods.
Example: Investment returns over 3 years
If you used the arithmetic mean instead: (20 + β10 + 15)/3 = 8.33% β which overstates the actual return because it ignores the compounding effect of losses.
The harmonic mean is the reciprocal of the arithmetic mean of reciprocals:
Harmonic Mean = n / (1/xβ + 1/xβ + β¦ + 1/xβ)
The harmonic mean is appropriate when averaging rates where the denominator is constant. The classic example is speed: if you drive 60 mph for the first half of a journey and 40 mph for the second half, your average speed is not (60+40)/2 = 50 mph.
The correct average speed (harmonic mean) = 2 / (1/60 + 1/40) = 2 / (0.0167 + 0.025) = 2 / 0.0417 = 48 mph.
The harmonic mean is also used in finance to calculate average price-earnings ratios across portfolios, and in physics for equivalent resistances in parallel circuits.
A moving average (or rolling average) smooths a time series by averaging over a sliding window of the most recent n data points. As new data arrives, the oldest point drops out and the newest enters.
A 3-period moving average of {10, 20, 30, 40, 50} produces: {β, β, 20, 30, 40}. The first two points have no moving average (insufficient data).
Moving averages are used extensively in stock price analysis (the 50-day and 200-day moving averages are standard technical indicators), in weather forecasting, in sales trend analysis, and in economics (GDP smoothing).
In the UK, the Office for National Statistics reports both mean and median household income. For 2024, the median UK household income was approximately Β£35,000 while the mean was considerably higher due to the effect of very high earners. The median is the "middle" value β 50% of households earn more, 50% earn less β and is much harder to distort with extreme outliers.
In the US, the Census Bureau similarly reports both. US median household income in 2023 was approximately $80,000, while the mean was higher. This difference is explained by income inequality: a small number of very high earners pull the mean up while the median is unaffected.
Which average is most "honest" for income? Neither β they tell different stories. The median tells you what a typical household earns. The mean tells you how national income per household would be divided if it were equally shared. Both are valid and important in different contexts.
In the UK, "average" at Key Stage 3 and GCSE typically means arithmetic mean, introduced through simple datasets. A-Level Statistics extends this to standard deviation, variance, the normal distribution, and measures of central tendency. The term "average" may also refer informally to median or mode in everyday language.
In the US, mean, median, and mode are introduced in middle school (grades 6β7) under the Common Core Standards. Advanced Placement Statistics (AP Statistics) covers sampling distributions, weighted averages, and regression β all of which build on the concept of the mean.
Mean is the arithmetic average (sum divided by count). Median is the middle value when data is sorted. Mode is the most frequent value. For income data, median is usually most representative because extreme high earners pull the mean up.
Use weighted average when different values have different levels of importance. Examples: grade calculations (exams weighted more than homework), portfolio returns (positions weighted by investment size), opinion polls (responses weighted by demographic representation).
Investment returns compound: losing 10% then gaining 10% leaves you 1% below where you started (0.90 Γ 1.10 = 0.99). The arithmetic mean would suggest 0% overall change. The geometric mean correctly accounts for compounding and gives the true average growth rate.
Use harmonic mean when averaging rates where the denominator is constant β particularly speeds, rates, and price-to-earnings ratios. For speed: if you travel equal distances at different speeds, the correct average speed is the harmonic mean of those speeds.
A moving average smooths out short-term fluctuations to reveal longer-term trends. In stock markets, crossing the 50-day MA above the 200-day MA is a bullish signal ("golden cross"). In economics, moving averages remove seasonal noise from monthly data.
No β the geometric mean requires all positive numbers (you cannot take the root of a negative product with real numbers). For investment returns, use growth factors (1 + return rate) which are always positive, then subtract 1 from the result.
For any set of positive numbers: HM β€ GM β€ AM (harmonic mean β€ geometric mean β€ arithmetic mean). Equality holds only when all values are identical. This is known as the AM-GM-HM inequality.
Standard deviation measures the average distance of each data point from the mean. A small standard deviation means values cluster closely around the mean. A large one means they are spread out. Together, mean and standard deviation describe the centre and spread of a distribution.
Results are for educational purposes. All calculations use standard mathematical formulas. Geometric and harmonic means require positive numbers only.