Proving simple inequalities
The simplest way to prove an inequality is the definition: to prove A ≥ B we show A − B ≥ 0, often with a perfect square. For example, a² + b² − 2ab = (a − b)² ≥ 0, so a² + b² ≥ 2ab. For two positive numbers a and b the arithmetic mean is (a + b) : 2 and the geometric mean is √(ab). The inequality (a + b) : 2 ≥ √(ab) holds because the difference is (√a − √b)² : 2 ≥ 0; equality happens only when a = b. With it we can show, for example, that x + 1/x ≥ 2 for x > 0 and find the smallest value of expressions whose product is constant. Another method is assuming the opposite: suppose the inequality fails and reach a contradiction (for example, a square being negative); then the original inequality is true.
A school website’s visitors grew 3 times in the first year and 12 times in the second. Overall that is 3 · 12 = 36 times. The equal yearly growth factor is √36 = 6, not the arithmetic mean (3 + 12) : 2 = 7.5: this is the inequality geometric mean ≤ arithmetic mean at work.
A financial analyst computes an average growth rate, a biologist estimates population growth, and a journalist reports growth figures correctly.
Suppose a value grew 5 times and then 20 times in two years: find the total growth, the equal yearly factor and the arithmetic mean, and say which is smaller.
“Proof chain”: teams prove an inequality such as a² + 9 ≥ 6a by the A − B ≥ 0 method and say when equality holds. Another team looks for mistakes.