Author(s)

Thomas Beatty, Gabriela von Linden

Department of Mathematics, Florida Gulf Coast University, Fort Myers, FL, USA.

Factoring quadratics over Z is a staple of introductory algebra and textbooks tend to create the impression that doable factorizations are fairly common. To the contrary, if coefficients of a general quadratic are selected randomly without restriction, the probability that a factorization exists is zero. We achieve a specific quantification of the probability of factoring quadratics by taking a new approach that considers the absolute size of coefficients to be a parameter n. This restriction allows us to make relative likelihood estimates based on finite sample spaces. Our probability estimates are then conditioned on the size parameter n and the behavior of the conditional estimates may be studied as the parameter is varied. Specifically, we enumerate how many formal factored expressions could possibly correspond to a quadratic for a given size parameter. The conditional probability of factorization as a function of n is just the ratio of this enumeration to the total number of possible quadratics consistent with n. This approach is patterned after the well-known case where factorizations are carried out over a finite field. We review the finite field method as background for our method of dealing with Z [x]. The monic case is developed independently of the general case because it is simpler and the resulting probability estimating formula is more accurate. We conclude with a comparison of our theoretical probability estimates with exact data generated by a computer search for factorable quadratics corresponding to various parameter values.

Factorization, Polynomial, Quadratic, Integers, Rational Numbers, Monic, Modular Arithmetic, Conditional Probability

Beatty, T. and Linden, G. (2020) On Conditional Probabilities of Factoring Quadratics. Advances in Pure Mathematics, 10, 114-124. doi: 10.4236/apm.2020.103008.