Linked from

The 68 pages that link to Least squares, each with the reason it gives.

• PolynomialRelated: Polynomial models are often fitted to data by minimizing squared errors.

• Carl Friedrich GaussRelated: Gauss used this method to refine astronomical measurements and determine orbits.

• Hilbert spaceRelated: In Hilbert spaces, least-squares solutions are projections onto subspaces of candidate models.

• VarianceRelated: Its squared-error objective connects model fitting to variance-based assumptions and estimates.

• TriangulationRelated: It reconciles small inconsistencies among redundant angles in a measured network.

• Euclidean spaceRelated: Squared Euclidean distances make the method a geometric projection problem.

• Linear algebraRelated: When equations are inconsistent, least squares finds the solution with the smallest residual error.

• Cauchy–Schwarz inequalityRelated: Inner-product bounds help establish projection formulas and error estimates in least-squares problems.

• Arithmetic meanRelated: For a single constant fitted to data, least squares selects the arithmetic mean.

• Inner productRelated: The normal equations express the residual's orthogonality to the model subspace.

• Inverse problemRelated: It supplies a common data-fit criterion, though alone it may not stabilize an ill-posed inversion.

• Dot productRelated: Normal equations use inner products to characterize the best-fitting solution.

• Quadratic formRelated: Its objective function is a quadratic form in the residual vector.

• Condition numberRelated: The design matrix’s condition number reveals sensitivity in fitted coefficients.

• Regression analysisRelated: Ordinary least squares fits coefficients by minimizing squared prediction errors.

• Linear equationRelated: When exact linear equations conflict with data, least squares finds a best-fit estimate.

• OrthogonalityRelated: At the minimum, the residual is orthogonal to the model’s column space.

• Geodetic datumRelated: Survey networks use it to estimate consistent coordinates and datum parameters from measurements.

• Kalman filterRelated: Under Gaussian assumptions, the filter’s estimate is also a least-squares solution.

• Orbit determinationRelated: Orbit fitting often adjusts parameters to minimize observation residuals.

• Orthogonal projectionRelated: Its fitted vector is the projection of the data onto a model's column space.

• Euclidean normRelated: The residual sum of squares is the squared Euclidean norm of the error vector.

• ResidualRelated: Its fitting criterion gives residuals a central role in estimating model parameters.

• Parameter estimationRelated: It estimates model parameters by minimizing residual error in regression and related models.

• Perpendicular linesRelated: In linear regression, residuals are orthogonal to the fitted subspace at the optimum.

• Inner product spaceRelated: Its residuals are orthogonal to the fitted model space at the minimum.

• TrilaterationRelated: With noisy or redundant ranges, it provides a common way to estimate the best-fitting position.

• Karl PearsonRelated: Pearson used least-squares reasoning in fitting curves to observed data.

• Kernel (linear algebra)Related: A nontrivial kernel means some parameter changes leave a model’s predictions unchanged.

• Rank (linear algebra)Related: Rank determines whether the least-squares solution is unique and whether columns are redundant.

• Pearson correlation coefficientRelated: In simple linear regression, correlation determines the fitted line’s slope after scaling.

• Direct methodRelated: QR factorization provides a direct route to least-squares solutions without forming normal equations.

• Scatter plotRelated: It commonly determines the regression line drawn through scatter-plot observations.

• Convex analysisRelated: Its convex quadratic objective makes it a standard model for convex optimization.

• Adjoint operatorRelated: The optimality condition for a linear least-squares problem uses the adjoint in the normal equations.

• Jan TinbergenRelated: Least-squares estimation helped fit Tinbergen’s equations to observed economic data.

• Affine subspaceRelated: Its fitted values lie in the column space, while residuals are measured from that linear subspace.

• Column spaceRelated: When a system is inconsistent, least squares projects its target onto the column space.

• Approximation theoryRelated: It produces best approximations in many finite-dimensional settings with a squared-error measure.

• Bessel's inequalityRelated: Orthogonal projection links best approximation errors to the squared-coefficient bound.

• Matrix theoryRelated: Matrix factorizations solve overdetermined data-fitting problems efficiently.

• Maximum and minimumRelated: Its fitted values are defined through a minimum of an error function.

• Ragnar FrischRelated: Econometric estimation connected Frisch's theoretical equations to observed economic measurements.

• Friedrich Wilhelm BesselRelated: Bessel used least-squares methods to combine observations and estimate astronomical quantities.

• Euclidean vectorRelated: Its geometric solution projects data onto a subspace using Euclidean inner products.

• Fundamental theorem of linear algebraRelated: The residual lies in the left null space of the coefficient matrix.

• Simon NewcombRelated: It provided a systematic way to fit astronomical theories to measurements.

• Apollonius's theoremRelated: The midpoint identity helps explain why averages minimize sums of squared distances.

• Dimensional metrologyRelated: It is commonly used to fit geometric features to measured points.

• History of statisticsRelated: Its early use in astronomy made error minimization a practical tool for extracting measurements.