Rough automatic draft, low quality: expect mistakes and awkward wording. It will be replaced by a better translation. Original →
37
The Case of the Method of Mathematical Induction
One day, a teacher gave 10th grade students the following homework problem:
To prove, by attempting to use the method of mathematical induction, that for any natural number n the inequality is true
Pn=2⋅4⋅6…2n1⋅3⋅5…(2n−1)<n+11.
Kolya V. found the following solution at home;
a) For n=1 the inequality is true, since
P1=21<21.
Let us show that the validity of inequality Pn+1<n+21 follows from condition Pn<n+11. To do this it is enough to show that
PnPn+1<n+2n+1,
or
2n+22n+1<n+2n+1,
or
(2n+1)2(n+2)<4(n+1)3.
or
4n3+12n2+9n+2<4n3+12n2+12n+4,
or
0<3n+2,
which is obviously true. Thus, a fairly provable inequality follows from the principle of mathematical induction.
However, while telling this solution at school, Kolya forgot to put one in the denominator and began to prove a more “gross” inequality:
Pn<n1.
Here’s what he got:
a) For n=1 the inequality is true, since
P1=21<11.
Let us show that condition Pn<n1 implies Pn+1<n+11. To do this it is enough to show that
PnPn+1<n+1n,
or
2n+22n+1<n+1n,
or
(2n+1)2(n+2)<4(n+1)3n,
or
4n3+12n2+9n+2<4n3+12n2+12n+4,
or
0<3n+2,
which is obviously true. Thus, a fairly provable inequality follows from the principle of mathematical induction.
For those who understand this issue, we propose to prove two inequalities:
Pn<3,1n1.
It is interesting that in this inequality the coefficient 3.1 cannot be replaced by 3.2. The largest coefficient at which it remains true is π (the ratio of the circumference to the diameter).