Mean Absolute Deviation Calculator

Measure the average distance of each data point from the mean. Visualize absolute deviations via dynamic bar chart — perfect for understanding data spread.

Supports decimals, negatives, up to 30 values for chart clarity.
? Simple: 4,8,6,5,3
? Exam scores: 78,85,92,88,79,84,90
? Finance returns: 5.2,-1.3,4.8,3.5,-2.1,6.0
? Wide spread: 2,18,25,4,30,7,19
⚖️ Zero deviation: 10,10,10,10
❤️ Blood pressure (mmHg): 118,122,119,121,120
Privacy-first: All calculations happen locally in your browser. No data is uploaded.

What is Mean Absolute Deviation (MAD)?

The Mean Absolute Deviation (MAD) is a robust measure of statistical dispersion. It represents the average absolute distance between each data point and the dataset's mean. Unlike variance or standard deviation, MAD is less sensitive to outliers and is expressed in the original unit, making interpretation intuitive.

MAD = (1/n) · Σ |xᵢ - μ| where μ = mean, n = number of observations.

Step-by-step calculation & visualization

  1. Compute the mean (μ) = sum of all values / count.
  2. Calculate absolute deviations |xᵢ - μ| for each point.
  3. Average those deviations → MAD.
  4. Bar chart displays each absolute deviation; the red dashed line marks the MAD value.

Real-world applications

? Financial risk

Portfolio daily returns: [0.5%, -0.2%, 0.8%, -0.4%, 0.3%]. MAD = 0.36% indicates low volatility. Used instead of standard deviation to reduce impact of extreme market moves.

? Clinical consistency

Blood pressure readings (mmHg): 118, 122, 119, 121, 120 → μ=120, MAD=1.2. Low MAD suggests reliable measurement protocol.

?️ Weather forecasting

Daily temperature errors (forecast vs actual): [2°C, -1°C, 3°C, -2°C, 1°C] → MAD = 1.8°C, a robust measure of forecast accuracy.

MAD vs. Standard Deviation – Mathematical relationship

Property Mean Absolute Deviation (MAD) Standard Deviation (σ)
Outlier sensitivity Low (linear penalty) High (quadratic)
Units Same as data Same as data
Interpretation Average absolute distance Root mean squared deviation
For normal distribution MAD ≈ 0.7979 × σ (Geary, 1935)

For normally distributed data, the constant 0.7979 provides a quick conversion: σ ≈ MAD / 0.7979 or MAD ≈ 0.8·σ for approximation.

Limitations of MAD (important nuance)

  • Not suitable for inferential statistics: Unlike variance, MAD lacks additive properties and is rarely used in hypothesis testing.
  • Skewed distributions: MAD from the mean may be misleading; consider median absolute deviation (MAD about median) for highly skewed data.
  • Less efficient for normal data: For Gaussian distributions, standard deviation has better statistical efficiency (lower standard error).

Our calculator focuses on mean absolute deviation, a foundational measure taught in introductory statistics and applied in exploratory data analysis.

MAD in Time Series Forecasting

Forecast error metrics: Mean Absolute Deviation of errors (also called MAE – Mean Absolute Error) is widely preferred over MSE because it penalizes all errors linearly. Example: Sales forecast errors [10, -15, 5, -8] → MAD = 9.5 units, directly interpretable as average mistake magnitude.

Expert-backed methodology — This tool is reviewed by getzenquery tech team and validated against R's mean(abs(x - mean(x))) and Python's NumPy. References: “Robust Statistics” by Huber (Wiley), NIST/SEMATECH e-Handbook, and Geary (1935) “The ratio of the mean deviation to the standard deviation as a test of normality”. Updated June 2026.

Frequently Asked Questions

MAD = 0 only when all data points are identical — no variability. Our chart handles this case by showing a horizontal line at zero and zero-height bars.

Absolutely. In forecasting, MAD of errors (MAE) is a standard metric. It is less sensitive to outliers than MSE and is expressed in original units.

Bars represent absolute deviations, so they are always ≥0. The chart's y-axis starts at zero, ensuring no negative bars. The mean reference line (blue dot) shows original mean location.

MAD uses the mean as reference; median absolute deviation uses the median and is even more robust to outliers. This calculator focuses on mean-based MAD, commonly taught in introductory statistics.
References: Wolfram MathWorld; NIST Handbook; Geary, R.C. (1935). "The Ratio of the Mean Deviation to the Standard Deviation".