Sequence - Part 3

This is the third part of our series on sequences. There will be no new material in this post. The main purpose of this post is to go through some examples so we can get used to the method of solving recurrence equations. If you have not read the previous parts of our series, please read them here: Part 1, Part 2.

Sequence - Part 2

This is the second part of our series on sequences. Please have a careful read of Part 1 before you continue to this post. Today, we will introduce some more terminologies about sequences and we will present a general method to solve a linear recurrence equation.

Sequence - Part 1

Today we will begin our first post of a series on recurrence sequences. The purpose of this series is to show you how to derive a general formula for a sequence that satisfies a linear recurrence equation. We start with some basic operations on sequences. We will learn about addition of sequences and multiplication of a sequence with a number.

Fibonacci sequence and Pascal triangle

In previous post, we used the result of a tile matching puzzle to prove an identity for Fibonacci sequence. Today, we will continue on and prove another identity. We will show that $${2011 \choose 0} + {2010 \choose 1} + {2009 \choose 2}+ {2008 \choose 3}+ \dots + {1007 \choose 1004}+ {1006 \choose 1005} = F_{2012},$$ $${2012 \choose 0} + {2011 \choose 1} + {2010 \choose 2}+ {2009 \choose 3}+ \dots + {1007 \choose 1005}+ {1006 \choose 1006} = F_{2013}.$$
In general, we have the following identity $$\sum_{v+u=n}{v \choose u} = F_{n+1}.$$ Through this identity, we will see an interesting connection between the Fibonacci sequence and the Pascal triangle.

Fibonacci identities

In previous post, we have learnt about Fibonacci sequence and a tile matching puzzle. Today, we will use the result of the tile matching puzzle to prove the following identity on Fibonacci sequence $$F_{n+m+2} = F_{n+1} F_{m+1} + F_{n+1} F_m + F_n F_{m+1}.$$

Like an art painting, a mathematical problem reflects different beauties if we view it at different angles. Take, for example, the Fibonacci identity that we are considering, we can prove this identity by different methods. One method is to use the formula $$F_n = \frac{1}{\sqrt{5}} \left[ \left( \frac{1 + \sqrt{5}}{2} \right)^n - \left( \frac{1 - \sqrt{5}}{2} \right)^n \right]$$ to calculate the two sides of the identity and make them equal by straightforward algebraic manipulations. Another method is to make use of the tile matching puzzle that we learned in the previous post. I hope that you will find this "combinatorial" method more interesting.

Fibonacci numbers and a tile matching puzzle

Today we will learn about Fibonacci sequence and a tile matching puzzle. We will soon see that the tile matching puzzle seems to be unrelated to the Fibonacci sequence, but it turns out that the solution to the puzzle is, in fact, the Fibonacci sequence!

Let us first introduce the Fibonacci sequence. The Fibonacci sequence $\{F_n\}$ is determined by the following rule: $$F_0 = 0, ~F_1 = 1, ~F_{n+1} = F_n + F_{n-1}.$$

Thus, $$F_0 = 0, ~F_1 = 1, ~F_2 = 1, ~F_3 = 2, ~F_4 = 3, ~F_5 = 5, ~F_6 = 8, ~F_7 = 13, ~F_8 = 21, \dots$$

de Moivre's formula

In previous post we have learned about complex numbers. Today, we will learn about the trigonometric form of a complex number and the famous de Moivre's formula.

Complex number

Today we will learn about the complex numbers. The crucial point about complex numbers is that we accept a very special number that we will denote it by $i$. This number $i$ is very special because it satisfies the following identity $$i^2 = -1.$$

So a complex number will have the form $$a + ib$$ where $a$ and $b$ are two real numbers. When $b=0$ then $a + ib = a$ is just a normal real number, and when $a=0$ then $a + ib = ib$ is called a pure imaginary number. Here are some examples of complex numbers: $$1+ i, ~~ 2 - 3i, ~~ -\sqrt{3} + 4i, ~~5i - 4, ~~6, ~~i, ~~-3i, ~~4 + 2i, \dots$$

A facebook friendship problem

Do you know that facebook has its own mathematical problem? It is called the facebook friendship problem. Never heard of it?!

Well, this problem is to prove that if you pick out 6 arbitrary people on facebook, then among these 6 people, you can find 3 people such that they are all mutually friend of each other, or there are no friendships among them at all. 

For example, in the figure below, on the left we have 6 people: Amy, Ben, Cindy, Dan, Emma and Francis. If two of them are friend on facebook then we connect them by a purple line, and if they are not friend on facebook then we connect them by a blue line. 

We can see that in this example, we can pick out 3 people, which are Amy, Dan and Emma, they are all mutually friend of each other because they are connected by purple lines. The same is true for Francis, Ben and Cindy.

In the example on the right, we have 6 people: Kate, Linda, Matt, Noah, Olivia and Peter. Among them we can pick out 3 people, which are Olivia, Noah and Matt, there are no friendships among them because they are connected by blue lines. The same is for Peter, Noah and Matt. 

Power sum and Wolstenholme's theorem

In the last post, we have learned how to derive formulas for the power summation $$S_k(n) = 1^k + 2^k + 3^k + \dots + n^k.$$

Today we will investigate some divisibility properties of the power summation $S_k(n)$. We will prove that if $p$ is a prime number and $k$ is not divisible by $p-1$ then $$S_k(p-1) = 1^k + 2^k + 3^k + \dots + (p-1)^k = 0 \pmod{p}.$$

We will also investigate the following summation $$S_{-k}(n) = \frac{1}{1^k} + \frac{1}{2^k} + \frac{1}{3^k} + \dots + \frac{1}{n^k}.$$

There is a theorem in number theory concerning this summation, it is called Wolstenholme's theorem.

Wolstenholme's Theorem. If $p$ is a prime number $>3$ then $$S_{-1}(p-1) = \frac{1}{1} + \frac{1}{2} + \frac{1}{3} + \dots + \frac{1}{p-1} ~=_{Q} ~0 \pmod{p^2} $$ and $$S_{-2}(p-1) = \frac{1}{1^2} + \frac{1}{2^2} + \frac{1}{3^2} + \dots + \frac{1}{(p-1)^2} ~=_{Q} ~0 \pmod{p}.$$

We will prove a general result, that is, if $p$ is a prime number and $k$ is not divisible by $p-1$ then $$S_{-k}(p-1) = \frac{1}{1^k} + \frac{1}{2^k} + \frac{1}{3^k} + \dots + \frac{1}{(p-1)^k} ~=_{Q} ~0 \pmod{p}.$$

Power sum

Today we will learn about how to derive the formula for power summation $$S_k(n) = 1^k + 2^k + 3^k + \dots + n^k .$$


Our main tool is the following binomial identity
$$(x+y)^k = x^k + {k \choose 1} x^{k-1} y + {k \choose 2} x^{k-2} y^2 + {k \choose 3} x^{k-3} y^3 + \dots + {k \choose {k-1}} x y^{k-1} + y^k.$$

Euclidean algorithm

In previous post, we have learned about Bézout's Lemma. Today, we will learn about Euclidean algorithm. This algorithm is used to determine the coefficients in the Bézout's equation.

Let us recall Bézout's lemma. 

Bézout's Lemma. Let $a$ and $b$ be two integers and $d$ their greatest common divisor. Then there exist two integers $x$ and $y$ such that $$d = a ~x + b ~y.$$

We use Euclidean algorithm to calculate the greatest common divisor $d$ of the two numbers $a$ and $b$, and determine the two values of $x$ and $y$ in Bézout's equation $$d = a ~x + b ~y.$$

We will see that Euclidean algorithm is motivated from a very simple and natural idea.