How Do You Find Relative Extrema of the Function h(x) = cos(π/x)?

  • Thread starter Thread starter k3k3
  • Start date Start date
  • Tags Tags
    Extrema
Join the discussion
Registration is free. Start your own thread to ask a follow-up.
3 replies · 2K views
k3k3
Messages
76
Reaction score
0

Homework Statement


Let h:(0,1)→ℝ be defined by h(x)=cos([itex]\pi\\[/itex]/x). Determine the set of all x such that x has a relative extreme value at x.



Homework Equations





The Attempt at a Solution


It is common knowledge that the min and max are -1 and 1. Using intuition, it is clear by guess and check to arrive at x=1/(2n+1) gives a minimum and x=1/2n gives a maximum.

Using the first derivative test, I arrive at 1=0 which is not true. I am unsure how to interpret this answer.

Trying to solve [itex]d^{2}[/itex](cos([itex]\pi\\[/itex]/x))/d[itex]x^{2}[/itex]=0 is proving to be a real challenge.

I am only concerned about this since my professor wants me to show my work, so I do not think saying I guessed and checked will be acceptable.
 
Physics news on Phys.org
k3k3 said:

Homework Statement


Let h:(0,1)→ℝ be defined by h(x)=cos([itex]\pi\\[/itex]/x). Determine the set of all x such that x has a relative extreme value at x.



Homework Equations





The Attempt at a Solution


It is common knowledge that the min and max are -1 and 1. Using intuition, it is clear by guess and check to arrive at x=1/(2n+1) gives a minimum and x=1/2n gives a maximum.

Using the first derivative test, I arrive at 1=0 which is not true. I am unsure how to interpret this answer.

Trying to solve [itex]d^{2}[/itex](cos([itex]\pi\\[/itex]/x))/d[itex]x^{2}[/itex]=0 is proving to be a real challenge.

I am only concerned about this since my professor wants me to show my work, so I do not think saying I guessed and checked will be acceptable.
Instead of guessing at the answer, find the values of x for which h'(x) = 0.

Show us what you did to arrive at 1 = 0.
 
Mark44 said:
Instead of guessing at the answer, find the values of x for which h'(x) = 0.

Show us what you did to arrive at 1 = 0.

Using the chain rule, h'(x)=[itex]\pi\\[/itex]sin([itex]\pi\\[/itex]/x)*1/x[itex]^{2}[/itex]

Set it equal to 0

[itex]\pi\\[/itex]sin([itex]\pi\\[/itex]/x)*1/x[itex]^{2}[/itex]=0

Multiply both sides by x[itex]^{2}[/itex]/[itex]\pi\\[/itex] yields sin([itex]\pi\\[/itex]/x)=0

Take the arcsin of both sides yields

[itex]\pi\\[/itex]/x=0

Divide both sides by pi,

1/x=0

Multiply both sides by x

1=0

Or am I forgetting that there are more possible values that arcsin can be 0?
 
k3k3 said:
Using the chain rule, h'(x)=[itex]\pi\\[/itex]sin([itex]\pi\\[/itex]/x)*1/x[itex]^{2}[/itex]

Set it equal to 0

[itex]\pi\\[/itex]sin([itex]\pi\\[/itex]/x)*1/x[itex]^{2}[/itex]=0

Multiply both sides by x[itex]^{2}[/itex]/[itex]\pi\\[/itex] yields sin([itex]\pi\\[/itex]/x)=0

Take the arcsin of both sides yields

[itex]\pi\\[/itex]/x=0

Divide both sides by pi,

1/x=0

Multiply both sides by x

1=0

Or am I forgetting that there are more possible values that arcsin can be 0?

You should not be looking at the arcsin; you should be looking for solutions of the equation sin(w) = 0. Can you see why the arcsin misses all but one solution?

RGV