linear regression

Straight-line statistical modeling

Linear regression is a statistical method for modeling the relationship between variables by fitting a straight line through data points.

The goal is prediction: given one variable, estimate another.

The linear regression showed that each additional year of education corresponded to higher earnings.

The technique assumes a roughly linear relationship and minimizes the distance between observed values and the fitted line.

evaluation

residuals

Ongoing compensation for rebroadcasts

confidence interval · outliers · leverage

methodology

least squares

Error-minimizing statistical method

intercept · best fit · assumptions · normality · linearity · independence · overfitting · underfitting · regularization · lasso

variables

independent variable

Factor deliberately manipulated

regressor · covariate · feature · bivariate · multivariate

correlation

correlation coefficient

Measure of variable association

relationship · association · covariance · strength · direction · positive correlation · negative correlation · causation · confounding variable

Categories
Opposites
modeling approach

linear regressionneural network

Interconnected computational nodes mimicking cognition

decision tree · deep learning

functional form complexity

linear regressioncurve fitting

Approximating data mathematically

methodological approach

linear regressionneural networks

Layered computational learning systems

decision trees · deep learning · boosting · bagging · classification

Family

line, linear, regress, regression, nonlinear, regressive, regressed, curvilinear, regresses

Familiarity
Top 500,000
Profile
Used worldwide · Timeless · Abstract · Rated G