
We refer to the maximum and minimum values as ______ ______ or ______ (singular: ______) and to the process of finding them as _________. Sometimes, we are interested in finding the min or max for x in a particular interval I, rather than on the entire domain of f
 extreme values or extrema
 singular: extremum
 optimization

Define: Extreme values on an Interval
Let f be a function on an interval I and let a ∈ I. We say that f(a) is the
 Absolute minimum of f on I if f(a) ≤ f(x) for all x ∈ I
 Absolute maximum of f on I if f(a) ≥ f(x) for all x ∈ I

Does every function have a minimum or maximum value? If not, state an example
Clearly not, as we see by taking f(x) = x. Indeed, f(x) increases without bound as x → ∞ and decreases without bound as as x → ∞

In fact, extreme values do not always exist even if we restrict ourselves to an interval I. Name two scenarios where things can go wrong
 Discontinuity: (Fig 2A pg 175) shows a discontinuous function with no maximum value. The values of f(x) get arbitrarily close to 3 from below, but 3 is not the maximum value because f(x) never actually takes on the value 3.
 Open interval: In (Fig 2B), g(x) is defined on the open interval (a,b). It has no max because it tends to ∞ on the right, and it has no min because it tends to 10 on the left without ever reaching this value.

Explain the Existence of Extrema on a Closed Interval Theorem
A continuous function f on a closed (bounded) interval I = [a,b] takes on both a minimum and a maximum value on I.

Define both cases of Local Extrema:
We say that f(c) is a
 Local minimum occurring at x = c if f(c) is the minimum value of f on some open interval (in the domain of f) containing c
 Local maximum occurring at x = c if f(c) is the maximum value of f on some open interval (in the domain of f) containing c

A local max occurs at x = c if (c, f(c)) is the highest point on the graph within some small box [Fig 3(A) pg 176]. Thus, f(c) is greater than or equal to all other nearby values, but it does not have to be the ______ ______ value of f.
absolute maximum

Local minima are similar. Fig 3B (pg 176) shows the difference between local and absolute extrema: f(a) is the absolute max on [a,b] but is not a local max. (Why?)
because f(x) takes on larger values to the left of x=a

What is the crucial observation when finding local extrema? (explain). What if f'(c) is not differentiable?
 The crucial observation is that the tangent line at a local min or max is horizontal [fig. 4A pg 176]. In other words, if f(c) is a local min or max, then f'(c) = 0.
 Then the tangent line may not exist

Define Critical Points
A number c in the domain of f is called a critical point if either f'(c) = 0 or f'(c) does not exist

Fermat's Theorem on Local Extrema
If f(c) is a local min or max, then c is a critical point of f

Extreme Values on a Closed Interval Theorem
Assume that f is continuous on [a,b] and let f(c) be the minimum or maximum value on [a,b]. Then c is either a critical point or one of the endpoints a or b

As an application of our optimization methods, we prove Rolle's Theorem: If f is differentiable between two points a and b, then somewhere between these two points, the derivative is ____. Graphically, what happens if the secant line between x = a and x = b is horizontal?
 zero
 At least one tangent line between a and b is also horizontal

Formally state Rolle's Theorem
Assume that f is continuous on [a,b] and differentiable on (a,b). If f(a) = f(b), then there exists a number c between a and b such that f'(c) = 0

Rolle's Theorem Proof: Since f is continuous and [a,b] is closed, f has a min and a max in [a,b]. Where do they occur? Specifically what happens if either the min or max occurs at a point c in the open interval (a,b)? What happens if the min and max occur at the endpoints and f(a) = f(b)?
 If the min or max occurs at a point c in the open interval, then f(c) is a local extreme value and f'(c) = 0 by Fermat's Theorem.
 If the min and max occur at the endpoints and f(a) = f(b), then the min and max coincide and f is a constant function with zero derivative. Then f'(c) = 0 for all c in (a,b)

Verify Rolle's Theorem for:
f(x) = x^{4}  x^{2} on [2,2]
Hint: Illustrating Rolle's Theorem
 The hypotheses of Rolle's Theorem are satisfied because f is differentiable (and therefore continuous) everywhere, and f(2) = f(2):
 f(2) = 2^{4} 2^{2} = 12, f(2) = (2)^{4}  (2)^{2} = 12
 We must verify that f'(c) = 0 has a solution in (2,2) so we solve f'(x) = 4x^{3} 2x = 2x(2x^{2}  1) = 0.
 The solution are c = 0 and c = ±1/√2 ~ ±0.707. They all lie in (2,2), so Rolle's Theorem is satisfied with three values of c

Show that f(x) = x^{3} + 9x  4 has precisely one real root
Hint: Use Rolle's Theorem
 First, we note that f(0) = 4 is negative and f(1) = 6 is positive.
 By the Intermediate Value Theorem (Section 2.8), f has at least one root a in [0,1].
 If f had a second root b, then f(a) = f(b) = 0 and Rolle's Theorem would imply that f'(c) = 0 for some c ∈ (a,b).
 This is not possible because f'(x) = 3x^{2} + 9 ≥ 9, so f'(c) = 0 has no solutions.
 We can conclude that a is the only real root of f (fig. 14)

