Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Let be $a, b$ in $(0,1)$ such that $a+b>1$. I need to prove that:


I'm looking for an elementary proof that doesn't resort to the calculus tools.

share|improve this question
Are you allow to use the fact that $2^a + 3^b$ is convex? Then you just need to check at the end-points $(a,b) \in \{ (0,1), (1,0), (1,1)\}$. –  Willie Wong Jun 15 '12 at 12:36
@Willie Wong: yes. This idea came to me too late. –  Chris's sis the artist Jun 15 '12 at 12:41

2 Answers 2

up vote 12 down vote accepted

From the graph of the function $f(x)=2^x$ we see that on interval $(0,1)$ it is bellow the line $y=x+1$ joining the points $(0,f(0))=(0,1)$ and $(1,f(1))=(1,2)$. Thus we have $$2^a<1+a$$ for $a\in(0,1)$.

Using similar argument for $3^x$ we get $$3^b<1+2b$$ for $b\in(0,1)$.

Adding the two inequalities together and using $1<a+b$ we obtain $$2^a+3^b<2+a+2b<2(a+b)+a+2b=3a+4b.$$

share|improve this answer
nice proof! Thanks! –  Chris's sis the artist Jun 15 '12 at 12:41

Using Bernoulli's Inequality, $2^a \leq 1 + a$ and $3^b \leq 1 + 2b$. Therefore, $$2^a + 3^b \leq 2 + a + 2b < 2(a + b) + a + 2b = 3a + 4b$$

share|improve this answer
Oh dear, just realized its the same as the above answer. embarassed –  TenaliRaman Jun 15 '12 at 14:03
The problem is that both proofs use calculus in a hidden way ;) Proving Bernoulli for $a \in (0,1)$ requires calculus, and so does the concavity of $2^x$.... –  N. S. Jun 15 '12 at 14:18
That is true, but I think convexity of $2^x$ can be established without calculus. $2^{\lambda x + (1 - \lambda)y} = (2^x)^{\lambda} (2^y)^{1 - \lambda} \leq \lambda 2^x + (1 - \lambda) 2^y$ where the last step comes from the generalized AM-GM inequality. –  TenaliRaman Jun 15 '12 at 16:54
True, but as far as I know, the generalized AM-GM inequality can only be proven for rational powers without Calculus ;) Anyhow, your point is probably as much non-calculus as possible for this problem... –  N. S. Jun 15 '12 at 19:55
@N.S. I was expecting some catch-22 like that :-). But then, we can use the fact that for f(x) continuous, it is enough to show midpoint convexity (thus using AM-GM in a rational power setting). Of course, to show f(x) is continuous we may have to resort to calculus again :P –  TenaliRaman Jun 15 '12 at 19:58

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.