MHB How can AM-GM be used to solve the Inequality Challenge II?

anemone
Gold Member
MHB
POTW Director
Messages
3,851
Reaction score
115
Prove that

$\sqrt[3]{\dfrac{2}{1}}+\sqrt[3]{\dfrac{3}{2}}+\cdots+\sqrt[3]{\dfrac{996}{995}}-\dfrac{1989}{2}<\dfrac{1}{3}+\dfrac{1}{6}+\cdots+ \dfrac{1}{8961}$
 
Mathematics news on Phys.org
anemone said:
Prove that

$\sqrt[3]{\dfrac{2}{1}}+\sqrt[3]{\dfrac{3}{2}}+\cdots+\sqrt[3]{\dfrac{996}{995}}-\dfrac{1989}{2}<\dfrac{1}{3}+\dfrac{1}{6}+\cdots+ \dfrac{1}{8961}$
Not a solution, but a possible line of approach:
[sp]Start by noticing that $\dfrac{1989}2 = 995 - \dfrac12$, and $\dfrac1{8961} = \dfrac1{9\cdot995+6}.$ Next, notice that $\dfrac12 = \dfrac13+\dfrac16$. So we can subtract $\dfrac12$ from the left side, and $\dfrac13+\dfrac16$ from the right side, so that the inequality becomes $$\sqrt[3]{\dfrac{2}{1}} +\sqrt[3]{\dfrac{3}{2}} +\ldots +\sqrt[3]{\dfrac{996}{995}} - 995 < \frac19 + \frac1{12} + \frac1{15} + \ldots + \frac1{9\cdot995} + \frac1{9\cdot995+3} + \frac1{9\cdot995+6}.$$ If we write this as $$\sum_{n=1}^{995}\Bigl(\sqrt[3]{\dfrac{n+1}n} - 1\Bigr) < \sum_{n=1}^{995}\Bigl(\frac1{9n} + \frac1{9n+3} + \frac1{9n+6}\Bigr),$$ then it is tempting to think that each term on the left might be less than the corresponding term on the right, in other words $$\sqrt[3]{\dfrac{n+1}n} - 1 < \frac1{9n} + \frac1{9n+3} + \frac1{9n+6}\quad (\text{for all }n\geqslant 1).$$ I believe that is true. It certainly holds for $n=1$, because $\sqrt[3]{2} -1 \approx 0.2599$ and $\frac19 + \frac1{12} + \frac1{15} \approx 0.2611$. It also appears to hold for other values of $n$. But it is a delicate inequality – the difference between the two sides is extremely small – and I have not found any convenient way to prove it.[/sp]

Edit, Oh, it's suddenly obvious!
[sp]It's just an AM-GM argument: $$\begin{aligned} \sqrt[3]{\dfrac{n+1}n} &= \sqrt[3]{\dfrac{3n+1}{3n} \cdot \dfrac{3n+2}{3n+1} \cdot \dfrac{3n+3}{3n+2}} \\ &< \frac13\Bigl( \dfrac{3n+1}{3n} + \dfrac{3n+2}{3n+1} + \dfrac{3n+3}{3n+2} \Bigr) \\ &= \frac13\Bigl(1 + \frac1{3n} + 1 + \frac1{3n+1} + 1 + \frac1{3n+2} \Bigr) \\ &= 1 + \frac1{9n} + \frac1{9n+3} + \frac1{9n+6}. \end{aligned}$$ [/sp]
 
Last edited:
Opalg said:
Not a solution, but a possible line of approach:
[sp]Start by noticing that $\dfrac{1989}2 = 995 - \dfrac12$, and $\dfrac1{8961} = \dfrac1{9\cdot995+6}.$ Next, notice that $\dfrac12 = \dfrac13+\dfrac16$. So we can subtract $\dfrac12$ from the left side, and $\dfrac13+\dfrac16$ from the right side, so that the inequality becomes $$\sqrt[3]{\dfrac{2}{1}} +\sqrt[3]{\dfrac{3}{2}} +\ldots +\sqrt[3]{\dfrac{996}{995}} - 995 < \frac19 + \frac1{12} + \frac1{15} + \ldots + \frac1{9\cdot995} + \frac1{9\cdot995+3} + \frac1{9\cdot995+6}.$$ If we write this as $$\sum_{n=1}^{995}\Bigl(\sqrt[3]{\dfrac{n+1}n} - 1\Bigr) < \sum_{n=1}^{995}\Bigl(\frac1{9n} + \frac1{9n+3} + \frac1{9n+6}\Bigr),$$ then it is tempting to think that each term on the left might be less than the corresponding term on the right, in other words $$\sqrt[3]{\dfrac{n+1}n} - 1 < \frac1{9n} + \frac1{9n+3} + \frac1{9n+6}\quad (\text{for all }n\geqslant 1).$$ I believe that is true. It certainly holds for $n=1$, because $\sqrt[3]{2} -1 \approx 0.2599$ and $\frac19 + \frac1{12} + \frac1{15} \approx 0.2611$. It also appears to hold for other values of $n$. But it is a delicate inequality – the difference between the two sides is extremely small – and I have not found any convenient way to prove it.[/sp]

That is a good line with all excellent observations, Opalg!(Nerd)

Opalg said:
Edit, Oh, it's suddenly obvious!
[sp]It's just an AM-GM argument – details to follow tomorrow.[/sp]

I can't wait to read your solution...:o
 
anemone said:
I can't wait to read your solution...:o
Previous comment now edited to complete solution. (Emo)
 
Opalg said:
Edit, Oh, it's suddenly obvious!
[sp]It's just an AM-GM argument: $$\begin{aligned} \sqrt[3]{\dfrac{n+1}n} &= \sqrt[3]{\dfrac{3n+1}{3n} \cdot \dfrac{3n+2}{3n+1} \cdot \dfrac{3n+3}{3n+2}} \\ &< \frac13\Bigl( \dfrac{3n+1}{3n} + \dfrac{3n+2}{3n+1} + \dfrac{3n+3}{3n+2} \Bigr) \\ &= \frac13\Bigl(1 + \frac1{3n} + 1 + \frac1{3n+1} + 1 + \frac1{3n+2} \Bigr) \\ &= 1 + \frac1{9n} + \frac1{9n+3} + \frac1{9n+6}. \end{aligned}$$ [/sp]

Bravo, Opalg! The solution that I have is quite similar as yours and thus I won't reveal it and for your information, I am not the mastermind who provided that solution. :o
 
Insights auto threads is broken atm, so I'm manually creating these for new Insight articles. In Dirac’s Principles of Quantum Mechanics published in 1930 he introduced a “convenient notation” he referred to as a “delta function” which he treated as a continuum analog to the discrete Kronecker delta. The Kronecker delta is simply the indexed components of the identity operator in matrix algebra Source: https://www.physicsforums.com/insights/what-exactly-is-diracs-delta-function/ by...
Fermat's Last Theorem has long been one of the most famous mathematical problems, and is now one of the most famous theorems. It simply states that the equation $$ a^n+b^n=c^n $$ has no solutions with positive integers if ##n>2.## It was named after Pierre de Fermat (1607-1665). The problem itself stems from the book Arithmetica by Diophantus of Alexandria. It gained popularity because Fermat noted in his copy "Cubum autem in duos cubos, aut quadratoquadratum in duos quadratoquadratos, et...
Thread 'Imaginary Pythagorus'
I posted this in the Lame Math thread, but it's got me thinking. Is there any validity to this? Or is it really just a mathematical trick? Naively, I see that i2 + plus 12 does equal zero2. But does this have a meaning? I know one can treat the imaginary number line as just another axis like the reals, but does that mean this does represent a triangle in the complex plane with a hypotenuse of length zero? Ibix offered a rendering of the diagram using what I assume is matrix* notation...
Back
Top