Editing Reduction (section)

=== Pure mathematics and statistics ===
<!--Please organize meanings with sub-bullets as applicable-->
* Reducible as the opposite of [[irreducible (mathematics)]]
* [[Reduction (mathematics)]], the rewriting of an expression into a simpler form
* [[Beta reduction]], the rewriting of an expression from lambda calculus into a simpler form
* [[Dimension reduction]], the process of reducing the number of random variables under consideration
* [[Lattice reduction]], given an integer lattice basis as input, to find a basis with short, nearly orthogonal vectors
* [[Subject reduction]] or preservation, a rewrite of an expression that does not change its type
* [[Reduction of order]], a technique for solving second-order ordinary differential equations
* [[Reduction of the structure group]], for a <math>G</math>-bundle <math>B</math> and a map <math>H \to G</math> an <math>H</math>-bundle <math>B_H</math> such that the pushout <math>B_H \times_H G</math> is isomorphic to <math>B</math>
* [[Reduction system]], reduction strategy, the application of rewriting systems to eliminate reducible expressions
* [[Reduced form]], in statistics, an equation which relates the endogenous variable X to all the available exogenous variables, both those included in the regression of interest (W) and the instruments (Z)
* [[Reduced homology]], a minor modification made to homology theory in algebraic topology, designed to make a point have all its homology groups zero
* [[Reduced product]], a construction that generalizes both direct product and ultraproduct
* [[Reduced residue system]], a set of φ(n) integers such that each integer is relatively prime to n and no two are congruent modulo n
* [[Reduced ring]], a ring with no non-zero nilpotent elements
* [[Reduced row echelon form]], a certain reduced row echelon form of a matrix which completely and uniquely determines its row space
* [[Reduced word]], in a free group, a word with no adjacent generator-inverse pairs
* [[Variance reduction]], a procedure used to increase the precision of the estimates that can be obtained for a given number of iterations