show that
\(d\mid n\) and \(c\mid(n/d)\) if and only if \(c\mid n\) and \(d\mid(n/c)\)
\(d\mid n\)->this means d is divisible by n right?
yes
\(n/d\) is the regular division
If we consider c=7 n=2 d=16 d is divisible by n and c/(n/d) = 56 c is not divisible by n but still the condition holds true
sorry i was wrong earlier
\(d\mid n\)->this means d is divisible by n right? this is wrong
\(d \mid n\) means \(d\) divides \(n\)
example : \(2\mid 6\) \(3\mid 21\)
when reading you may replace "\(\mid\)" with "divides"
\[ \frac{ n }{ d} \div c =(n \div c) \div d\]
thats right, but how is that relevant here ?
the Rhs has to be a whole number so (n/c) is whole or c|n and n/c should also be divisible by d, or d|(n/c)
ohk.. got you :)
ok :) \(\Large\frac{ n }{ d }=a\) and \(\Large\frac{ n }{ dc }=b\) \(\large{a~and~b}\) are integers now we look at this-> \(\Large\frac{ n }{ dc }=b\) We know that d will divide n but then b is integer so c has to divide n or we can also say c has to divide n/d we had to prove \(c\mid n\) which is done and \(d\mid(n/c)\) is same as \(\Large\frac{ n }{ dc }=b\) :)
Nice!
This is part of a very beautiful proof.. I can share the proof if you're interested... because my next questions are based on this and I don't want to attempt them alone :)
yes, do share :)
yes share \(\Huge\ddot\smile\)
i see you're very good with \(\tau\) and \(\sigma\) functions
there is another function that is needed for the theorem that i am talking about
It is called mobius mu function : \(\mu\) function please see the attached quick for the definition of this function
examples : 1) \(\mu(3^2*5)=0\) because \(3^2\) divides \(3^2*5\) 2) \(\mu(7)=(-1)^1 = -1\) 3) \(\mu(7*11) = (-1)^2 = 1\)
it is a very simple function, but is also very powerful
Ok :)
just to understand \(\mu\) function better, maybe lets attempt that problem
okay (:
we still on main problem ? or have another question ?
ok for a part the number n will be either divisible or 4 or it won't be if \(\mu(n)\) is divisible by 4 or we can say \(2^2\) then its proved because then \(\mu(n)=0\) if the number is not divisible by 4 it will be of the form 4k+1 or 4k+2 or 4k+3 its sure that if the n is of this ^ form them one of (n+1) / (n+2) / (n+3) will be divisible by 4 so for that \(]mu(n)\) will be 0
you're using the fact that one integer will be divisible by 4 in a set of four consecutive integers. nice
yes :)
main question is solved @ikram002p
that was easy, part b can be a bit tricky..
ok so it will be like this-\[\mu(1!)+\mu(2!)+\mu(3!)+\mu(4!)........\] \(\mu(4!)=0\) because it is divisible by 4 all the \(\mu(n!)\) after \(\mu(4!)\) will be 0 because its sure they will be divisible by 4 so all the part from \(\mu(4!)\) becomes 0 we are left with this-\(\mu(1!)+\mu(2!)+\mu(3!)\) this is equal to 1+(-1)^1 +(-1)^2 =1
well part two we only need to know that \(\mu (n)=0 ~if~ p^2|n \) and we know for n>=4, 2^2|n! so \(\Large \sum_{k=1}^n \mu(k!)=\\\Large \sum_{k=1}^4 \mu(k!)= \mu(1!)+\mu(2!)+\mu(3!)+\mu(4!)=1+(-1)+(1)+0=1\)
oh always late xD
lol
well i just not i put 4 instead of 3 =D . but nothing would change.
\(\mu(n!) =0\) for n>=4 so it should be \(\Large \sum_{k=1}^n \mu(k!)= \Large \sum_{k=1}^3 \mu(k!)\\= \Large\mu(1!)+\mu(2!)+\mu(3!) \\\Large =1+(-1)+(1) =1\)
suppose \(g(n) = \sum\limits_{d\mid n}f(d)\), is there a way to find the value of \(f(n)\) ?
if gn is defined yes there is, think of a question so we could try.
Join our real-time social learning platform and learn together with your friends!