11.4 Three Principles
Three Methods
Principle of Mathematical Induction (PMI)
Principle of Complete Induction (PCI)
Principle of Well Ordering
- Useful for proving theorems of the form \(\left(\forall n)P(n\right)\).
- Definition of functions \(f(n)\).
Principle of Mathematical Induction
Let \(S \subseteq \mathbb{N}\) be a subset such that:
-
\(1 \in S\).
-
\(\forall n \, (n \in S \Rightarrow n+1 \in S)\).
Then \(S = \mathbb{N}\).
Example
-
Why \(17 \in S\)?
-
\(1 \in S \Rightarrow 1 \in S \Rightarrow 2 \in S\).
-
then \(2 \in S\).
-
\(2 \in S \Rightarrow 2 \in S \Rightarrow 3 \in S\).
-
then \(3 \in S\).
- \(3 \in S \Rightarrow 3 \in S \Rightarrow 4 \in S\).
So \(4 \in S\).... \(17 \in S\).
-
Particular definition
In general, a subset \(S \subseteq \mathbb{R}\) is called inductive if \(\forall x \, (x \in S \Rightarrow x+1 \in S)\).
For example, \(\{5,6,7,8,\ldots\}\) is inductive.
\(S = \{2, 4, 6, 8, \dots \}\) is not inductive because \(2 \in S \Rightarrow 3 \in S\) is false.
For example, \([1,\infty)\) is inductive.
Fact: \(\mathbb{N}\) is the smallest inductive subset of \(\mathbb{R}\) that contains \(1\).
Proof of \(\forall n \, P(n)\) by using the PMI
To show that \(\forall n \, P(n)\) is true, it is enough to show:
-
\(P(1)\) is true.
-
\(P(n) \Rightarrow P(n+1)\) for all \(n\).
Why? Let \(S\) be the true set of \(P(n)\).
- By (1), \(1 \in S\), so \(P(1)\) is true.
- Suppose \(n \in S\), so \(P(n)\) is true.
- By (2), \(P(n) \Rightarrow P(n+1)\) is true, so \(P(n+1)\) is true.
So \(n+1 \in S\). By the PMI, \(S = \mathbb{N}\).
Example
Prove that the sum of the first \(n\) odd numbers is \(n^2\) for all \(n\). \(P(n): 1 + 3 + 5 + \dots + (2n - 1) = n^2\)
Proof:
-
Base Case: We first prove \(P(1)\).
For \(n = 1\), we have:
- Left-Hand Side (L.H.S) \(= 1\)
- Right-Hand Side (R.H.S) \(= 1^2 = 1\)
So the result is true in this case. (We proved that \(P(1)\) is true)
-
Inductive Step: Assume that \(1 + 3 + \dots + (2n - 1) = n^2\) (Inductive Hypothesis, I.H.).
We want to show:
\(1 + 3 + \dots + (2n - 1) + 2(n + 1) - 1 = (n + 1)^2\)Now,
\(\text{L.H.S} = (1 + 3 + \dots + (2n - 1)) + 2(n + 1) - 1\)Using the I.H., this becomes:
\(= n^2 + 2n + 1 = (n + 1)^2 = \text{R.H.S}\)
Therefore, the result is true for all \(n\) by induction.
Remark
To show \(P(n) \Rightarrow P(n+1)\) in general, we use the direct method.
So we assume \(P(n)\) is true. This assumption is called the inductive hypothesis (IH).
We use the IH, \(P(n)\), and previous knowledge to deduce \(P(n+1)\).
After proving \(P(1)\) and \(P(n) \Rightarrow P(n+1)\), we can say: "The result is true for all \(n\) by induction."
Remark
If we want to show \((\forall n \geq 17) , P(n)\), it is enough to show:
-
\(P(17)\) is true.
-
\((\forall n \geq 17 ), (P(n) \Rightarrow P(n+1))\).
PMI and Recursive Definitions
To define a function \(f(n)\) for all \(n\), it is enough to:
- Define \(f(1)\).
- Define \(f(n+1)\) in terms of \(f(n)\).
Why? Let \(S = \{ n \in \mathbb{N} : f(n) \text{ is defined} \}\). Check that \(S = \mathbb{N}\) using PMI.
Example
Definition of \(n!\) (n factorial).
-
\(1! = 1\)
-
\((n+1)! = n! \cdot (n+1)\)
For example, \(10! = 9! \cdot 10 = 8! \cdot 9 \cdot 10 = 7! \cdot 8 \cdot 9 \cdot 10 = \dots = 1 \cdot 2 \cdot 3 \cdot 4 \cdot 5 \cdot 6 \cdot 7 \cdot 8 \cdot 9 \cdot 10\)
Definition of \(a^n\).
-
\(a^1 = a\)
-
\(a^{n+1} = a^n \cdot a\)
\(\sum\) and \(\prod\). Let \((a_n)\) be a sequence of real numbers.
We define \(\sum_{k=1}^n a_k\) (sum from 1 to \(n\)) as follows:
- \(\sum_{k=1}^1 a_k = a_1\)
- \(\sum_{k=1}^{n+1} a_k = \sum_{k=1}^n a_k + a_{n+1}\)
We define \(\prod_{k=1}^n a_k\) (product from 1 to \(n\)) as follows:
- \(\prod_{k=1}^1 a_k = a_1\)
- \(\prod_{k=1}^{n+1} a_k = \prod_{k=1}^n a_k \cdot a_{n+1}\)
For example, \(n! = \prod_{k=1}^n k\).
Example
Prove that \(\prod_{i=1}^n (4i - 2) = \frac{(2n)!}{n!}\) for all \(n\).
Proof:
For \(n = 1\):
- Left-Hand Side (LHS) \(= 4 \cdot 1 - 2 = 2\).
- Right-Hand Side (RHS) \(= \frac{(2 \cdot 1)!}{1!} = \frac{2!}{1!} = \frac{2 \cdot 1}{1} = 2\).
Thus, the result is true for \(n=1\).
We now assume \(\prod_{i=1}^n (4i - 2) = \frac{(2n)!}{n!}\) (Inductive Hypothesis, I.H.). We want to prove \(\prod_{i=1}^{n+1} (4i - 2) = \frac{(2(n+1))!}{(n+1)!}\)
Left-Hand Side (LHS):
\(\prod_{i=1}^{n+1} (4i - 2) = \left( \prod_{i=1}^n (4i - 2) \right) \cdot (4(n+1) - 2)\) \(= \frac{(2n)!}{n!} \cdot (4n + 2)\)
Right-Hand Side (RHS):
\(\frac{(2n+2)!}{(n+1)!} = \frac{(2n+2)(2n+1)(2n)!}{(n+1) \cdot n!}= \frac{2(2n+1) \cdot (2n)!}{n!} = (4n+2) \cdot \frac{(2n)!}{n!}\)
Thus, \(\text{LHS} = \text{RHS}\).
Therefore, the result is true for all \(n \in \mathbb{N}\).
PCI (Principle of Complete Induction)
If \(S \subseteq \mathbb{N}\) satisfies:
-
\(1 \in S\)
-
\(\forall n \left( \left( \forall k \leq n, k \in S \Rightarrow P(k) \right) \Rightarrow P(n+1) \right)\)
Then \(S = \mathbb{N}\).
Proof of \(\forall n, P(n)\) by using the PCI.
It is enough to prove two things:
-
\(P(1)\) is true.
-
\(\forall n \geq 1 \left( \left( \forall k \leq n, P(k) \right) \Rightarrow P(n+1) \right)\)
Why? Let \(S = \{ n : P(n) \text{ is true} \}\). Check that \(S = \mathbb{N}\) by using the PCI.
Remark:
To show that \(\forall n \geq 5, P(n)\) is true, it is enough to show:
-
\(P(5)\) is true.
-
\(\forall n \geq 5 \left( \forall 5 \leq k \leq n, P(k) \Rightarrow P(n+1) \right)\)
Example
Every natural number \(n \geq 2\) has a prime divisor.
Let \(P(n): n\) has a prime divisor.
Proof:
If \(n = 2\), \(2\) is a prime number and \(2 \mid 2\), so the result is true.
Let \(n \geq 2\), and assume that the result is true for every \(k\) such that \(2 \leq k \leq n\). We want to show the result is true for \(n+1\) (Inductive Hypothesis, I.H.).
- If \(n+1\) is prime, then \(n+1\) is a prime divisor of \(n+1\), and we are done.
- Otherwise, \(n+1 = a \cdot b\) with \(2 \leq a, b \leq n\).
By the I.H., \(a\) has a prime divisor \(p\). But then \(p\) is also a prime divisor of \(n+1 = a \cdot b\).
Hence, the result is true for all \(n \geq 2\).
PCI and Recursive Definition
To define \(f(n)\) for every \(n \in \mathbb{N}\), it is enough to:
-
Define \(f(1)\), and
-
Define \(f(n+1)\) in terms of \(f(1), f(2), \dots, f(n)\).
Why? Let \(S = \{ n \in \mathbb{N} : f(n) \text{ is defined} \}\). Check that \(S = \mathbb{N}\) by using the PCI.
Example
Fibonacci Sequence
Define the sequence \(f_1, f_2, f_3, \dots\) as follows:
- \(f_1 = 1\)
- \(f_2 = 1\)
- \(f_{n+1} = f_n + f_{n-1}\)
For example:
- \(f_3 = f_2 + f_1 = 1 + 1 = 2\)
- \(f_4 = f_3 + f_2 = 2 + 1 = 3\)
- \(f_5 = f_4 + f_3 = 3 + 2 = 5\)
Theorem: \(f_n = \frac{1}{\sqrt{5}} \left( \alpha^n - \beta^n \right)\)
where \(\alpha = \frac{1 + \sqrt{5}}{2}\) and \(\beta = \frac{1 - \sqrt{5}}{2}\).
Observation: \(\alpha, \beta\) are roots of \(x^2 - x - 1 = 0\).
- \(\alpha + \beta = 1\) and \(\alpha \beta = -1\)
- \(\alpha^2 = \alpha + 1\), \(\beta^2 = \beta + 1\)
- \(\alpha - \beta = \sqrt{5}\)
Proof:
For \(n = 1\), \(f_1 = 1\) and \(\frac{1}{\sqrt{5}} (\alpha - \beta) = \frac{\sqrt{5}}{\sqrt{5}} = 1\). So the result is true.
Now we show that \(\forall n \geq 1, \left( \forall 1 \leq k \leq n, P(k) \right) \Rightarrow P(n+1)\).
For \(n=1\), we show this conditional by proving that \(P(2)\) is true.
We have \(f_2 = 1\) by definition. On the other hand,
\(\frac{1}{\sqrt{5}} (\alpha^2 - \beta^2) = \frac{1}{\sqrt{5}} (\alpha - \beta)(\alpha + \beta) = \frac{\sqrt{5}}{\sqrt{5}} \cdot 1 = 1\),
so the conditional is true for \(n = 1\).
We now show the condition for \(n\geq 2\)
We assume that the result is true for \(k\) such that \(1\leq k\leq n\)
We want to show the result for \(n+1\), that is, \(f_{n+1} = \frac{1}{\sqrt{5}} \left( \alpha^{n+1} - \beta^{n+1} \right)\).
Left-Hand Side (LHS):
\(f_n + f_{n-1} = \frac{1}{\sqrt{5}} \left( \alpha^n - \beta^n \right) + \frac{1}{\sqrt{5}} \left( \alpha^{n-1} - \beta^{n-1} \right)\)\(=\frac{1}{\sqrt5}\left(\alpha^{n-1}\cdot\alpha-\beta^{n-1}\cdot\beta\right)+\frac{1}{\sqrt5}\left(\alpha^{n-1}-\beta^{n-1}\right)\)
\(=\)\(\frac{1}{\sqrt{5}} \left[ \alpha^{n-1} (\alpha + 1) - \beta^{n-1} (\beta + 1) \right]\)\(= \frac{1}{\sqrt{5}} \left( \alpha^{n-1} \alpha^2 - \beta^{n-1} \beta^2 \right)\)\(= \frac{1}{\sqrt{5}} \left( \alpha^{n+1} - \beta^{n+1} \right) = \text{R.H.S.}\)
So the result is true for \(n+1\).
Therefore, the result is true for all \(n\) by complete induction.
Principle of Well-Ordering
Every non-empty subset \(S \subseteq \mathbb{N}\) has a first (smallest) element.
We can use this principle to prove theorems of the form \(\forall n, P(n)\) by contradiction.