Home » Mathematics Statistics and Analysis Calculators » Data Scaler

Data Scaler

Use this Data Scaler to standardize and normalize your numerical data. Choose from multiple scaling methods, including Z-score, Min-Max, Median IQR, and Max-Abs. Import CSV files, visualize before/after results, and export processed data.

Scaling Methods & Formulas

1. Z-score Standardization 
Transforms data to have mean = 0 and standard deviation = 1.

Standard Deviation (σ):
  Zᵢ = (Xᵢ - μ) / σ
  μ = mean(X)
  σ = sqrt( Σᵢ (Xᵢ - μ)² / (n - 1) )

Median Absolute Deviation (MAD):
  Zᵢ = (Xᵢ - median(X)) / MAD(X)
  median(X) = 50th percentile of X
  MAD(X) = median(|Xᵢ - median(X)|)
2. Range Scaling (Min–Max) 
Scales data to a specified range [a, b]:

  Xscaledᵢ = a + (b - a) * (Xᵢ - min(X)) / (max(X) - min(X))
  Default range: [0, 1]
  min(X), max(X) = minimum and maximum values of the feature (column)
3. Median IQR (Robust Scaling) 
Centers and scales data using the median and interquartile range:

  Xscaledᵢ = (Xᵢ - median(X)) / IQR(X)
  median(X) = 50th percentile of X
  IQR(X) = Q₃ − Q₁ (75th percentile − 25th percentile)
  Qp = p-th percentile of X
  Percentiles are computed using the Hazen method: p(k) = (k − 0.5) / n
4. Max-Abs Scaler 
Scales each feature by its maximum absolute value:

  Xscaledᵢ = Xᵢ / max(|X|)
  Resulting values lie in the range of [-1, 1]
5. Scale (Division Methods) 
By first element:
  Xscaledᵢ = Xᵢ / X₁

By numeric scalar s:
  Xscaledᵢ = Xᵢ / s

By standard deviation σ:
  Xscaledᵢ = Xᵢ / σ

By MAD:
  Xscaledᵢ = Xᵢ / MAD(X)

By IQR:
  Xscaledᵢ = Xᵢ / IQR(X)
6. Center (Subtraction Methods) 
By numeric scalar s:
  Xcenteredᵢ = Xᵢ - s

By median:
  Xcenteredᵢ = Xᵢ - median(X)

By mean μ:
  Xcenteredᵢ = Xᵢ - μ

Rating: 4.8/5 (23 votes)