Definition
Linear regression describes how a dependent variable (yield, price, consumption) depends on one or more explanatory variables using a line or plane estimated by least squares. The coefficients say how much the outcome changes per unit change of an input; R² gives the share of variability the model explains.
Regression & models
The model relies on assumptions: a linear relationship, independent residuals with constant variance and roughly normal distribution, and no strong multicollinearity among predictors. Verify them with residual plots; outliers and influential points (Cook's distance) can distort the coefficients substantially.
With many predictors, stepwise selection or model comparison by adjusted R² and information criteria helps. Regression is also a prediction tool — report a prediction interval with the point estimate; it is always wider than the confidence interval of the mean.
In Statistica
Statistics → Multiple Regression covers simple and multiple regression, stepwise selection, diagnostics of residuals, influential points and multicollinearity (VIF, tolerance) and prediction of new values with intervals. Nonlinear relationships are handled by Nonlinear Estimation, and General Linear Models add categorical predictors.
Related terms
- CorrelationCorrelation measures the strength and direction of a linear relationship between two quantities with a number from −1 to 1.
- Logistic regressionLogistic regression models the probability of a binary outcome — a patient responds to treatment or not, a customer…
- Nonlinear regressionNonlinear regression fits data with a model that is nonlinear in its parameters — exponential growth, a logistic curve…
- Generalized linear models (GLM)Generalized linear models extend linear regression to outcomes that are not normally distributed: binary (logistic…
- OutliersAn outlier is an observation that differs conspicuously from the rest — a measurement or entry error, a different population, or a genuine rare event.
Knowledgebase guides
FAQ
- What does R² = 0.6 mean?
- That the model explains 60 % of the variability of the dependent variable. The remaining 40 % is due to omitted influences and chance. Use adjusted R² to compare models with different numbers of predictors.
- Can I include a categorical variable in regression?
- Yes, via indicator (dummy) variables — Statistica creates them automatically in General Linear Models.
- What is multicollinearity?
- Strong correlation among predictors; coefficients become unstable and hard to interpret. Spot it with VIF > 10 and fix it by dropping or merging variables (PCA).
Try it on your own data
Statistica free for 30 days
Full version, no credit card. Or get a pricing estimate in a minute.
Regression & models
Updated: September 2026.