Logistic Regression Calculator

Free logistic curve calculator. Fit S-curve growth y = L/(1+e^(-k(x-x0))) with step-by-step solutions, R², and graphs. Private browser-based.

Logistic Regression Visualization

Logistic Regression Calculator

Free logistic curve calculator. Fit S-curve growth y = L/(1+e^(-k(x-x0))) with step-by-step solutions, R², and graphs. Private browser-based.

Enter your data points

# X Y
Leave blank to auto-estimate

Results

Equation

Growth Rate (k)

Midpoint (x₀)

Carrying Capacity (L)

Predicted Y

Statistics

Statistic Value
Standard Error
Sample Size (n)
Degrees of Freedom

Chart

Step-by-Step Solution

How to Use This Logistic Regression Calculator

S-Curve Modeling

Fits logistic growth curves y = L / (1 + e^(-k(x - x0))) to your data.

Enter X & Y Data

Input paired values and optionally set the carrying capacity L.

Growth Parameters

Get growth rate k, midpoint x₀, carrying capacity L, and R².

Best For

Population saturation, product adoption, learning curves, and epidemiological models.

Logistic curve fitting requires 0 < Y < L. The carrying capacity L is auto-estimated from your data but can be overridden.

What Is Logistic Curve Fitting?

📐 Logistic curve fitting models growth processes that start exponentially but eventually slow down as they approach a maximum limit called the carrying capacity (L). The classic formula is y = L / (1 + e−k(x − x₀)), where k controls the steepness of the curve and x₀ is the midpoint (the x-value where y reaches L/2).

📊 Real-world examples include: (1) Population biology — bacterial colonies that slow as nutrients run out, (2) Marketing — product adoption that saturates as market share fills, (3) Epidemiology — infection curves that flatten as herd immunity builds, and (4) Machine learning — training accuracy that plateaus as the model converges.

📐 Logistic curves are S-shaped (sigmoid). They are fundamentally different from exponential curves because they have an upper bound. This makes them essential whenever unlimited growth is physically impossible.

How Logistic Curve Fitting Works

  • 1
    Determine L: Identify the carrying capacity — the maximum possible Y value. The calculator auto-estimates L as max(Y) × 1.05, but you can override it.
  • 2
    Linearize via logit: Transform each data point using z = ln(y / (L − y)). This converts the logistic model into a linear equation z = −k·x + k·x₀.
  • 3
    Perform linear regression: Apply ordinary least squares to (x, z). The slope equals −k and the intercept equals k·x₀.
  • 4
    Recover parameters: Compute k = −slope and x₀ = intercept / k. The final equation is y = L / (1 + e−k(x − x₀)).
  • 5
    Assess fit on original scale: Compute R² using the original (non-transformed) data to verify the S-curve actually fits the observations.

When to Use Logistic Curve Fitting

  • Growth has a natural upper limit or saturation point
  • Early data shows exponential-like growth that eventually slows
  • You need to estimate carrying capacity, midpoint, or growth rate
  • Modeling population dynamics, product adoption, or learning curves

When to Avoid Logistic Curve Fitting

  • When Y values can exceed any reasonable upper bound
  • When growth is strictly exponential without any slowing
  • When data is monotonically decreasing without an asymptote
  • When you have fewer than 5 data points (unreliable parameter estimates)

Frequently Asked Questions

What is logistic curve fitting?

📐 Logistic curve fitting estimates the S-shaped curve y = L / (1 + e−k(x − x₀)) from your data. It models growth that accelerates initially, then slows as it approaches a maximum limit L.

How is the carrying capacity L determined?

📊 The calculator auto-estimates L as max(Y) × 1.05 — slightly above your highest observed value. You can also enter your own L if you know the theoretical limit from domain knowledge.

What is the difference between logistic and exponential regression?

📊 Exponential regression assumes unlimited growth (y = a·ebx). Logistic regression adds a carrying capacity L that growth cannot exceed. Use logistic when you expect saturation; use exponential when you do not.

Related Regression Calculators

Discover more specialized regression modeling tools.

Need a different statistical model? Try our regression equation calculator with steps for linear and polynomial analysis.