Hans Walser, [20260525]

Binet's Formula

1     What it's about

Generalization of Binet's formula.

2     Problem

Given a sequence fn with the generalized Fibonacci recursion

 

(1)       fn = p fn–1 + q fn–2

 

and the arbitrary starting values ​​ f1 and f2.

We look for an explicit formula for calculating fn as a function of n.

3     Solution

We define:

 

(2)       c1 = (p + √(p2 + 4q))/2,          c2 = (p – √(p2 + 4q))/2

 

These are the two solutions of the quadratic equation c2p c ­– q = 0 for c.

We make the following approach:

 

(3)       fn = a c1n + b c2n

 

We therefore form a linear combination of the two geometric sequences c1n and c2n, both of which satisfy recursion (1).

To calculate the coefficients a and b, we use the initial values ​​​​ f1 and f2. We obtain the linear system of equations:

 

(4)       f1 = a c11 + b c21

            f2 = a c12 + b c22

 

We substitute the solutions a and b into (3). This is the explicit formula we are looking for.

4     Examples

For p = 1, q = 1, f1 = 1, and f2 = 1, we get the classic Fibonacci sequence (Table 1).

 

n

fn

1

1

2

1

3

2

4

3

5

5

6

8

7

13

8

21

9

34

10

55

Tab. 1: Classical Fibonacci sequence

For p = 1, q = –1, f1 = 3, and f2 = 4, we get a periodic sequence (Table 2). It is antiperiodic with a period of 3 and periodic with a period of 6.

 

n

fn

1

3

2

4

3

1

4

–3

5

–4

6

­–1

7

3

8

4

9

1

10

–3

Tab. 2: Periodic sequence

For p = i, q = 1, f1 = 3, and f2 = 4, we get a periodic sequence with complex numbers (Table 3). It is antiperiodic with a period of 6 and periodic with a period of 12. Furthermore, fn+3 = ifn.

 

1

3

2

4

3

3 + 4i

4

3i

5

4i

6

–4 + 3i

7

–3

8

–4

9

–3 – 4i

10

–3i

11

–4i

12

4 – 3i

13

3

14

4

15

3 + 4i

16

3i

17

4i

18

–4 + 3i

19

–3

20

–4

 Tab. 3: Complex values