My last post talked a lot about the Golden Ratio and how we can see and derive it from many different sources. This post will focus on a special formula for finding the nth Fibonacci number that involves the Golden Ratio, named Binet's formula for Jacques Philippe Marie Binet, who derived the formula (although sources indicate that Abraham de Moivre and Daniel Bernoulli had known about it beforehand).
To start, let's express the formula itself:
where is the Golden Ratio, and is its conjugate. They are both solutions to the equation
To make things clearer, I want to write the polynomial in a different way:
We can multiply any amount of s to the equation, and it will still preserve equality (0 times anything is 0, while the relative powers of the polynomial are preserved). This means that we can rewrite the quadratic as
or:
Since we haven't changed the validity of the equation, the solutions to the above polynomial are still and . This means that
This is meaningful in that the powers of and themselves satisfy the Fibonacci recurrence. So for any sequence defined as
also satisfies the Fibonacci recurrence:
So, we can choose some and such that it follows the Fibonacci sequence: namely,
solutions of which make sure that and fulfill the first two terms of the sequence. With some manipulation, we find that
and
So, with our arbitrary sequence , we can plug our back in to find a formula for the Fibonacci sequence itself:
Notice how we can use any value for here to find the nth term of any sequence that behaves like the Fibonacci sequence but with different starting numbers. All we'd have to do is find an that satisfies the system of equations
We do not even have to be limited to Fibonacci sequences. Since our understanding of this formula comes from the quadratic
why shouldn't there exist some formula that would tell us the nth Pell number, or for that matter, the nth term of a given Lucas sequence?
We can start with the Pell numbers, of which growth is analogous to the silver ratio as the Fibonacci numbers are to the golden ratio. Once again, we can start with our 'silver ratio' quadratic:
.
Once again, simply multiplying by throughout yields
For the silver ratio and its conjugate , then,
Once again, any sequence satisfied by the Pell relation will also satisfy
Thus, we need to solve the system of equations
Which yields
and we are left with the formula
For any Lucas sequence of the form
and a corresponding quadratic equation
we can derive a similar formula for the nth term.