Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Proof using mathematical induction

  1. Sep 6, 2015 #1
    Prove that n^5 - 5n^3 + 4n is divisible by 120. for every natural number n greater or equal to 3.

    First, i checked if it works for n=3 and it does,

    so i could assume it works for some k>=3 so i could write k^5 - 5k^3 + 4k as 120*a a is natural number

    so for k+1 i have:

    (k+1)^5 - 5(k+1)^3 + 4(k+1)

    and after applying Binomial theorem i got:

    k^5 - 5k^3 + 4k + 5k^4 + 10k^3 - 5k^2 - 10k =

    120a + 5k^4 + 10k^3 - 5k^2 - 10k

    so now i only have to prove that 5k^4 + 10k^3 - 5k^2 - 10k is divisible by 120 but the problem is that i don't know how, doing induction again would be too complicated, so i tried to see what happen if k is even and what happens if k is odd number, if it's even i can write it as 2 times some number b but i end up with

    80b^4 + 80b^3 - 20b^2 - 20b which is divisible by 20 and that doesn't mean is that is divisible by 120. What can i do, is this good approach?
     
  2. jcsd
  3. Sep 6, 2015 #2

    PeroK

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Re another approach: what about factorisation?
     
  4. Sep 6, 2015 #3
    I will have product of five consecutive natural numbers which is divisible by 120, but i have to use induction to prove this.
     
  5. Sep 6, 2015 #4

    mathman

    User Avatar
    Science Advisor
    Gold Member

    You don't need induction. 120=2.3.4.5. 5 consecutive positive numbers are divisible by 2, 3, 4, and 5. To clarify about even numbers - there will be 2 even numbers in the string, one of which will be divisible by 4.
     
  6. Sep 6, 2015 #5

    Mark44

    Staff: Mentor

    It would be useful (and maybe even required) for the OP to prove this statement.
     
  7. Sep 7, 2015 #6
    OK, i don't need induction to prove this, there are different ways, i know. But the problem statement is such that i have to use induction to prove it, otherwise it doesn't count, so how can i solve it using induction?
     
  8. Sep 7, 2015 #7

    PeroK

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    You rejected the idea of multiple inductions for no good reason that I can see. If you must use induction, go back to your original approach and work on the second polynomial you generated. It's got a factor of 5, which should make things easier.
     
  9. Sep 7, 2015 #8

    mathman

    User Avatar
    Science Advisor
    Gold Member

    [itex]5k^4+10k^3-5k^2-10k=5(k-1)k(k+1)(k+2)[/itex]. You can work it out from here.
     
  10. Sep 12, 2015 #9

    Actually i tried using induction more than once and i ended up with doing induction five times but i solved it, thanks for help.
     
  11. Sep 12, 2015 #10

    mathwonk

    User Avatar
    Science Advisor
    Homework Helper

    maybe technically induction is required, but isn't it obvious that in any sequence of 5 consecutive natural numbers, at least one is divisible by 5, one is divisible by 3 and one is divisible by 4, and another is divisible by 2? bingo. but i guess the point is to practice induction.
     
  12. Sep 12, 2015 #11

    FactChecker

    User Avatar
    Science Advisor
    Gold Member

    Factoring would also have worked for the first step. n5 - 5n3 + 4n = (n-2)(n-1)n(n+1)(n+2)
     
  13. Sep 13, 2015 #12

    Exactly, that is the point. Practicing mathematical induction.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Proof using mathematical induction
  1. Mathematical induction (Replies: 5)

  2. Mathematical Induction (Replies: 9)

  3. Mathematical induction (Replies: 2)

Loading...