synapticz

6.4 Least Squares and Data Fitting

Module 6 — Orthogonality, Projections and Least Squares. Least Squares and Data Fitting. Progressive conceptual, computational and reasoning questions with worked explanations.

by Synapticz Media
Oct 20, 2026
7 quizzes
Public
FREE

Explore this collection

Browse the topics and quizzes below. Expand or collapse any branch to focus on a subject.

Quiz Book Content

7 quizzes

Showing 1–7 of 7 quizzes · Page 1 of 1

6.4-01 Least squares as a closest-point problem

6.4-01 Least squares as a closest-point problem

Least squares as a closest-point problem Least squares finds the reachable output closest to a target when exact equations may be inconsistent. This set emphasizes observation-space geometry, the meaning of the objective, and what changes when the available model space grows. What you will practise Interpret squared residual objectives. Compute a simple constant fit. Explain existence and closest-point uniqueness. Separate fitting error from coefficient size and prediction quality. Begin with recognition and interpretation, then work through calculations, error diagnosis and more demanding reasoning. Single-select and multi-select questions ask you to evaluate the stated assumptions as well as the result. Worked explanations connect the answer to a method or a counterexample, so you can check why tempting alternatives fail. Before you start Review column spaces and orthogonal projection. Keep a written calculation or short justification for each response, especially when several statements can be correct. This microlearning supports building and comparing linear data-fitting models within Module 6 of the Complete Linear Algebra Course.

linear algebra
FREE
Start Quiz
6.4-02 A best-fit straight line

6.4-02 A best-fit straight line

A best-fit straight line Build and solve best-fit straight-line models from small data sets. The questions connect intercept and slope calculations with centered inputs, residual conditions, prediction, rank deficiency and the distinction between vertical and perpendicular error. What you will practise Construct a line-design matrix. Solve for intercept and slope. Check fitted predictions and residual constraints. Handle repeated inputs and through-origin models. Begin with recognition and interpretation, then work through calculations, error diagnosis and more demanding reasoning. Single-select and multi-select questions ask you to evaluate the stated assumptions as well as the result. Worked explanations connect the answer to a method or a counterexample, so you can check why tempting alternatives fail. Before you start Review normal equations and basic algebra. Keep a written calculation or short justification for each response, especially when several statements can be correct. This microlearning supports polynomial and multi-feature regression within Module 6 of the Complete Linear Algebra Course.

linear algebra
FREE
Start Quiz
6.4-03 Polynomial and basis-function fitting

6.4-03 Polynomial and basis-function fitting

Polynomial and basis-function fitting A model may be nonlinear in its input while remaining linear in its unknown coefficients. Practice builds design rows for polynomials and fixed basis functions, checks rank, compares equivalent bases and examines transformed-response objectives. What you will practise Evaluate monomial and nonmonomial design columns. Recognize linear versus nonlinear parameter dependence. Use distinct inputs to assess polynomial rank. Distinguish original-scale and transformed-scale fitting. Begin with recognition and interpretation, then work through calculations, error diagnosis and more demanding reasoning. Single-select and multi-select questions ask you to evaluate the stated assumptions as well as the result. Worked explanations connect the answer to a method or a counterexample, so you can check why tempting alternatives fail. Before you start Review straight-line least squares and polynomial algebra. Keep a written calculation or short justification for each response, especially when several statements can be correct. This microlearning supports flexible linear models without losing control of the fitting objective within Module 6 of the Complete Linear Algebra Course.

linear algebra
FREE
Start Quiz
6.4-04 Multiple linear regression

6.4-04 Multiple linear regression

Multiple linear regression Multiple linear regression combines several features into a single design matrix. This set focuses on dimensions, coefficient meaning and prediction, then examines feature dependence, category coding, units and the limits of causal interpretation. What you will practise Construct rows with intercept and multiple features. Interpret conditional model coefficients. Detect redundant feature columns. Check residual equations and unit conversions. Begin with recognition and interpretation, then work through calculations, error diagnosis and more demanding reasoning. Single-select and multi-select questions ask you to evaluate the stated assumptions as well as the result. Worked explanations connect the answer to a method or a counterexample, so you can check why tempting alternatives fail. Before you start Review linear least squares and column independence. Keep a written calculation or short justification for each response, especially when several statements can be correct. This microlearning supports interpretable multi-feature models and rank-aware fitting within Module 6 of the Complete Linear Algebra Course.

linear algebra
FREE
Start Quiz
6.4-05 Weighted least squares

6.4-05 Weighted least squares

Weighted least squares Weighted least squares changes the relative cost of observation errors. Through explicit calculations and structural questions, this set explains positive weights, square-root row scaling and the geometry underlying weighted normal equations. What you will practise Form and solve weighted fitting problems. Convert positive weights to a scaled ordinary problem. Check weighted residual conditions. Investigate common rescaling, zero weights and invalid negative weights. Begin with recognition and interpretation, then work through calculations, error diagnosis and more demanding reasoning. Single-select and multi-select questions ask you to evaluate the stated assumptions as well as the result. Worked explanations connect the answer to a method or a counterexample, so you can check why tempting alternatives fail. Before you start Review ordinary least squares and diagonal matrices. Keep a written calculation or short justification for each response, especially when several statements can be correct. This microlearning supports modeling unequal observation priorities with clear mathematical assumptions within Module 6 of the Complete Linear Algebra Course.

linear algebra
FREE
Start Quiz
6.4-06 Uniqueness and rank deficiency

6.4-06 Uniqueness and rank deficiency

Uniqueness and rank deficiency A unique best fitted vector can coexist with infinitely many coefficient vectors. This microlearning develops that distinction using duplicate columns, null-space families, minimum-norm selection and near-dependence examples. What you will practise Identify the correct uniqueness criterion. Describe all coefficient minimizers. Separate exact dependence from sensitivity. Preserve fitted outputs when removing redundant columns. Begin with recognition and interpretation, then work through calculations, error diagnosis and more demanding reasoning. Single-select and multi-select questions ask you to evaluate the stated assumptions as well as the result. Worked explanations connect the answer to a method or a counterexample, so you can check why tempting alternatives fail. Before you start Review rank–nullity and the closest-point theorem. Keep a written calculation or short justification for each response, especially when several statements can be correct. This microlearning supports reliable interpretation of rank-deficient regression within Module 6 of the Complete Linear Algebra Course.

linear algebra
FREE
Start Quiz
6.4-07 Checking fit quality

6.4-07 Checking fit quality

Checking fit quality Assess a fit by examining its residuals and the meaning of its error measure. The questions combine error calculations with structured residuals, held-out checks, scale effects and the limitations of numerical optimality alone. What you will practise Calculate squared error and a stated RMSE. Interpret patterns in residuals. Distinguish training fit from generalization. Understand what normal-equation checks establish. Begin with recognition and interpretation, then work through calculations, error diagnosis and more demanding reasoning. Single-select and multi-select questions ask you to evaluate the stated assumptions as well as the result. Worked explanations connect the answer to a method or a counterexample, so you can check why tempting alternatives fail. Before you start Review least-squares residuals and basic squared-error arithmetic. Keep a written calculation or short justification for each response, especially when several statements can be correct. This microlearning supports responsible evaluation of fitted models within Module 6 of the Complete Linear Algebra Course.

linear algebra
FREE
Start Quiz