There is a single sequence of integers \( a_2, a_3, a_4, a_5, a_6, a_7\) such that
\(\frac{5}{7} = \frac{a_2}{2!} + \frac{a_3}{3!} + \frac{a_4}{4!} + \frac{a_5}{5!} + \frac{a_6}{6!} + \frac{a_7}{7!},\)
and \(0 \le a_i < i \) for \(i = 2, 3, \ldots, 7.\) Find \(a_2 + a_3 + a_4 + a_5 + a_6 + a_7.\)
I have been stuck on this problem for awhile.
So I figured that in order to solve this problem, I need to get all the denominators the same value. I have been puzzled on how to achieve ths however.
The first thing we do is move the a7/7! to the other side:
\(\frac57-\frac{a_7}{7!}=\frac{5\cdot6!}{7!}-\frac{a_7}{7!}=\frac{5\cdot6!-a_7}{7}\frac1{6!},\)
we require the first fraction to be an integer. This is the case for a7=2, now we can bring the next fraction to the left:
\(\frac{514}{6!}-\frac{a_6}{6!}=\frac{514-a_6}{6}\frac{1}{5!},\)
for the first fraction to be an integer a6 needs to equal 4. We bring the next fraction to the left:
\(\frac{85}{5!}-\frac{a_5}{5!}=\frac{85-a_5}{5}\frac1{4!},\)
which requires a5 to be zero. We bring the next fraction to the left:
\(\frac{17}{4!}-\frac{a_4}{4!}=\frac{17-a_4}{4}\frac1{3!},\)
which requires a4 to be 1. We continue on our course:
\(\frac{4}{3!}-\frac{a_3}{3!}=\frac{4-a_3}3\frac12,\)
obviously a3 should be one. We are left with:
\(\frac12=\frac{a_2}{2!},\)
and we get a2=1, case solved. Checking our answer with the calculator:
\(\frac{1}{2!}+\frac{1}{3!}+\frac{1}{4!}+\frac{4}{6!}+\frac{2}{7!}-\frac57=0.0000000000000000143,\)
which is purely a machine error, since 1/7! equals 0.0001984126984127 which is far larger than our error.
Note that 5/7 = 0.7142857 [ the 7142857 is a repeating string ]
Try this :
a2 = 1, a3 = 1, a4 = 1, a5 = 0, a6 = 4, a7 = 2
1/2! + 1/3! + 1/4! + 0/5! + 4/6! + 2/7! = 5 / 7
I didn''t really have a mathematical way to solve this.....but, note that
1/2! + 1/3! + 1/4! + 1/5! = 43/60 > 5/7
And
1/2! + 1/3! + 1/4! = 17/24 < 5/7
So......we need that 5/7 - 17/24 = 1/168 = a6/6! + a7/7!
So........it was just a case of letting a5 = 0 and then manipulating a6 and a7 to find a solution
Perhaps some other mathematician can find a more elegant way to the solution....!!!!