1

Let $gcd(a,b)=(a,b), lcm(a,b)=[a,b]$. if $n=2$, $$ab=(a,b)[a,b]$$ $n=3$, $$\frac{(a,b,c)^2}{(a,b)(b,c)(c,a)}=\frac{[a,b,c]^2}{[a,b][b,c][c,a]}$$ I know its proof. Let $$a=\prod_{i=1}^{s}p_i^{\alpha_i},b=\prod_{i=1}^{s}p_i^{\beta_i},c=\prod_{i=1}^{s}p_i^{\gamma_i}$$ For $1\leq i\leq s$, $$2max(\alpha_i,\beta_i,\gamma_i)+\sum_{cyc}{min(\alpha_i,\beta_i)}=2min(\alpha_i,\beta_i,\gamma_i)+\sum_{cyc}{max(\alpha_i,\beta_i)}$$Since this equation is symmetric, we may assume that $\alpha_i\geq\beta_i\geq\gamma_i$ and prove easily.

But I can't extend that for $n$ integers. It has strange logic. I found if $n=4$, $$\sum_{sym}max(\alpha_i,\beta_i,\gamma_i)+\sum_{sym}min(\alpha_i,\beta_i)=\sum_{sym}min(\alpha_i,\beta_i,\gamma_i)+\sum_{sym}max(\alpha_i,\beta_i)$$ $n=5$, $$2\sum_{sym}max(\alpha_i,\beta_i,\gamma_i)+3\sum_{sym}min(\alpha_i,\beta_i)=2\sum_{sym}min(\alpha_i,\beta_i,\gamma_i)+3\sum_{sym}max(\alpha_i,\beta_i)$$ But $n=6$, $$\sum_{sym}max(\alpha_i,\beta_i,\gamma_i)+2\sum_{sym}min(\alpha_i,\beta_i)\neq\sum_{sym}min(\alpha_i,\beta_i,\gamma_i)+2\sum_{sym}max(\alpha_i,\beta_i)$$ Can anyone help me?

Mutse
  • 571
  • Could you rigorously define what you want? For example, I could multiply all the pairwise GCDs and pairwise LCMs and say that this is trivially equal to the product of all the numbers raised to the power $n-1$. I assume this isn't what you are expecting... – Haran May 21 '20 at 09:39
  • @Haran I want to extend $n=3$ for all $n$.(not only pairwise GCDs and LCMs) – Mutse May 21 '20 at 11:04
  • So you want all possible GCDs and LCMs of all subsets (with atleast size $2$) of the $n$ numbers? – Haran May 21 '20 at 11:36
  • @Haran No, I want to extend at least 2 GCDs or LCMs respectively(not all posible). – Mutse May 21 '20 at 11:47
  • 1
    You have phrased your question in a very unclear manner. What do you mean by expression of GCDs and LCMs? Please clarify which GCDs and LCMs you want. What do you mean at least 2 GCDs or LCMs? – Haran May 21 '20 at 11:53
  • @Haran I'm sorry. I mean I want to use GCDs and LCMs like $n=3$. There is no need to use all possible GCDs and LCMs(in $n=4$, I didn't use $(a,b,c,d),[a,b,c,d]$), but not only pairwise GCDs and LCMs.. – Mutse May 21 '20 at 12:38
  • See also here and here and their Linked questions. – Bill Dubuque May 21 '20 at 15:18

0 Answers0