STATISTICS

Regression Calculators

Fit a line, polynomial curve, multiple-predictor model or association measure to observations. Review the equation, observed points, model errors and assumptions. Every calculator accepts pasted data or a local CSV/TSV file.

AVAILABLE CALCULATORS

Fit and inspect a model

The result includes model metrics and row-level residuals so you can inspect how the fit behaves across the input observations.

01Linear regression calculatorStraight-line equation, scatter plot, R², RMSE and residualsAvailable 02Quadratic regression calculatorDegree-two curve, vertex, R², RMSE and residualsAvailable 03Exponential regression calculatorLog-space least squares, positive Y, original-scale metricsAvailable 04Logarithmic regression calculatorFit Y against ln(X), positive X, equation and graphAvailable 05Polynomial regression calculatorDegree selector 1–5, QR solve, conditioning warnings and residualsAvailable 06Pearson correlation calculatorScatter plot and r with association-not-causation noteAvailable 07Multiple linear regression calculatorOLS with intercept, QR solve, scaling/rank warnings and conditional coefficientsAvailable 08Equation ExplorerPlot expressions with one to three variables, or upload CSV sample pointsAvailable

Choose a model shape

Use a linear model when one straight line is a reasonable summary of the relationship. Use a quadratic model when the observations show one bend. The polynomial calculator lets you compare degrees 1–5, but a higher R² alone does not justify a more complex model. Exponential and logarithmic models are useful when the theoretical relationship or residual pattern suggests a multiplicative or logarithmic form. Inspect residuals and validate predictions on data not used to fit it.

Linear regression · Quadratic regression · Polynomial regression