Skip to main content

Section 1.4 Recurrence Relations

Recurrence relations may be review or may be a new definition for you. Let me give the definition.

Definition 1.4.1.

A sequence \(\{a_n\}_{n=1}^\infty\) of real numbers is called a linear recurrence relation if each term \(a_n\) is a linear function of the previous \(k\) terms \(a_{n-1}, \ldots a_{n-k}\text{.}\) \(k\) is called the order of the recurrence relation. Since terms depends on previous terms, the first \(k\) terms of the sequence must be explicitly defined.
The most famous example is the Fibonacci sequence, which is a second order linear recurrence relation. Its terms, \(f_n\text{,}\) start with \(f_1 = f_2 = 1\) and then obey the linear recurrence \(f_n = f_{n-1} + f_{n-2}\text{.}\)
If I am given a recurrence relation relation, I don’t have a direct way to calculate the term \(a_n\) without calculating all the previous terms. Solving a recurrence relation means finding a fixed formula \(a_n = f(n)\) that describes the \(n\)th term of the sequence. These can be easy or difficult to find. The fixed formula for Fibonnaci is
\begin{equation*} f_n = \frac{(1+\sqrt{5})^n - (1-\sqrt{5})^n}{2^n \sqrt{5}}\text{.} \end{equation*}
The coefficients of a taylor series are sequences of real numbers. In the use of taylor series in this course, sequences of coefficients will be presented as recurrence relations. I will want to solve the recurrence relations to give a closed formula for the coefficients, in order to understand the resulting taylor series. Though there is a whole mathematical theory of recurrence relation, in this course I will solve them mostly by inspection —by looking at the terms and trying to determine a pattern directly. I won’t rely on other techniques which are likely unfamiliar to you.