If \( \alpha \) is a root of a polynomial equation \( P(x) = 0 \), then \( P(\alpha) = 0 \), and if \(M\) is the maximum coefficient value of \( |a_{n}|, |a_{n-1}|, |a_{n-2}|, \cdots ,|a_2|, |a_1| \) then \( nM \ge |a_{n}| + |a_{n-1}| + |a_{n-2}| + \cdots + |a_2| + |a_1| \). One of the absolute value inequalities can be used effectively for proving polynomial equation properties, \( |x + y| \le |x| + |y| \). You can find the proof of absolute value inequalities.
Worked Example
Let \( \alpha \) be a root of \( P(x) = x^n +A_{n-1} x^{n-1} + A_{n-2} x^{n-2} + \cdots + A_2 x^2 + A_1 x + A_0 \).
Let \( M \) be the maximum value of \( |A_{n}|, |A_{n-1}|, |A_{n-2}|, \cdots , |A_2|, |A_1| \).
(a) Show that \( |\alpha |^n \ge M(|\alpha|^{n-1} + |\alpha|^{n-2} + \cdots + |\alpha| + 1) \), given \( |a + b| \ge |a| + |b| \).
\( \begin{aligned} \require{AMSsymbols} \require{color} \displaystyle
P(\alpha) &= 0 &\color{red} \text{since } \alpha \text{ is a root} \\
0 &= \alpha^n + A_{n-1} \alpha^{n-1} + A_{n-2} \alpha^{n-2} + \cdots + A_1 \alpha + A_0 \\
\alpha^n &= -A_{n-1} \alpha^{n-1}-A_{n-2} \alpha^{n-2}-\cdots-A_1 \alpha-A_0 \\
|\alpha^n| &= |-A_{n-1} \alpha^{n-1}-A_{n-2} \alpha^{n-2}-\cdots-A_1 \alpha-A_0| \\
|\alpha|^n &= |A_{n-1} \alpha^{n-1} + A_{n-2} \alpha^{n-2} + \cdots + A_1 \alpha + A_0| \\
|\alpha|^n &\le |A_{n-1} \alpha^{n-1}| + |A_{n-2} \alpha^{n-2}| + \cdots + |A_1 \alpha| + |A_0| &\color{red} |x + y| \le |x| + |y|\\
&= |A_{n-1}||\alpha^{n-1}| + |A_{n-2}||\alpha^{n-2}| + \cdots + |A_1||\alpha| + |A_0| \\
&\le M|\alpha^{n-1}| + M|\alpha^{n-2}| + \cdots + M|\alpha| + M \\
&= M(|\alpha^{n-1}| + |\alpha^{n-2}| + \cdots + |\alpha| + 1) \\
&\le M(|\alpha|^{n-1} + |\alpha|^{n-2} + \cdots + |\alpha| + 1) &\color{red} |\alpha^n| \le |\alpha|^n \\
\therefore |\alpha|^n &\le M(|\alpha|^{n-1} + |\alpha|^{n-2} + \cdots + |\alpha| + 1)
\end{aligned} \)
(b) Show that \(1 < | \alpha| < 1 + M \) for any root \( \alpha \) of \( P(x) \).
\( \begin{aligned} \displaystyle \require{AMSsymbols} \require{color}\ \newcommand\ddfrac[2]{\frac{\displaystyle #1}{\displaystyle #2}}
|\alpha|^n &\le (|\alpha|^{n-1} + |\alpha|^{n-2} + \cdots + |\alpha| + 1) \\
|\alpha|^n &\le M \times \frac{|\alpha|^n-1}{|\alpha|-1} &\color{red} a + ar + ar^2 + \cdots + ar^{n-1} = \frac{a(r^2-1)}{r-1} \\
1 &\le M \times \frac{1-\ddfrac{1}{|\alpha|^n}}{|\alpha|-1} &\color{red}\text{divide both sides by } |\alpha|^n \\
1 &\le M \times \frac{1-\ddfrac{1}{|\alpha|^n}}{|\alpha|-1} < \frac{1}{|\alpha|-1} \\
1 &< \frac{1}{|\alpha|-1} \\
(|\alpha|-1)^2 &< M \times (|\alpha|-1) &\color{red}\text{multipy both sides by } (|\alpha|-1)^2 \\
(|\alpha|-1)^2-M \times (|\alpha|-1) &<0 \\
(|\alpha|-1)( |\alpha|-1-M ) &< 0 \\
\therefore 1 &< |\alpha| < 1 + M
\end{aligned} \)
Algebra Algebraic Fractions Arc Binomial Expansion Capacity Common Difference Common Ratio Differentiation Double-Angle Formula Equation Exponent Exponential Function Factorials Factorise Functions Geometric Sequence Geometric Series Index Laws Inequality Integration Kinematics Length Conversion Logarithm Logarithmic Functions Mass Conversion Mathematical Induction Measurement Perfect Square Perimeter Prime Factorisation Probability Product Rule Proof Quadratic Quadratic Factorise Ratio Rational Functions Sequence Sketching Graphs Surds Time Transformation Trigonometric Functions Trigonometric Properties Volume
