4.1.28 · D4Calculus I — Limits & Derivatives

Exercises — Applications — increasing - decreasing, local extrema (first derivative test)

2,818 words13 min readBack to topic
Figure — Applications — increasing - decreasing, local extrema (first derivative test)

The figure above is the sign-chart tool we lean on throughout — study it once here so every solution below can just say "build the sign chart." The recipe it pictures: mark critical points, split the line into intervals, sample one test point per interval, record the sign of there. A sign flip is a hilltop; is a valley bottom. It is drawn for the exact critical points and of L2·Q2 below (signs ), so read that solution against this picture.


Level 1 — Recognition

Recall Solution

What we do: plug the test values into and read the sign. We do not need itself — the sign of is the entire map of rising/falling.

  • decreasing at .
  • increasing at .
Recall Solution

Two ways to be critical: or undefined while is in the domain of .

  • : the fraction is zero only when its top is zero → . ✓
  • undefined: at the denominator is zero. But is not in the domain of , so it is not a critical point of — a point must belong to the domain to count.

Answer: the only critical point is .


Level 2 — Application (run the recipe)

Recall Solution

Step 1 — differentiate. (power rule term by term; the constant has slope ). Step 2 — critical points. . is a polynomial, defined everywhere, so no "undefined" points. Step 3–4 — test points either side of :

  • : decreasing on .
  • : increasing on . Step 5 — classify. Sign goes local minimum at . Its value is . Answer: decreasing on , increasing on ; local min at .
Recall Solution

Step 1. . Step 2. Set to zero: . Step 3–4. Three intervals; test one point each (this is exactly the chart in the top figure):

  • : → increasing on .
  • : → decreasing on .
  • : → increasing on . Step 5.
  • At : local max, value .
  • At : local min, value . Answer: local max , local min .

Level 3 — Analysis (undefined derivatives, subtle traps)

Recall Solution

Step 1. . Step 2 — critical points. has no solution (the numerator is never zero). But is undefined at (division by ), and is in the domain of (since ). So is a critical point — the "derivative-does-not-exist" kind. Step 3–4. Test either side. The key is the sign of : it is negative for and positive for .

  • : , so → decreasing on .
  • : , so → increasing on . Step 5. Sign local minimum at , value . Geometrically this is a sharp cusp — the graph comes down, points, and goes back up. Answer: local min at ; no other critical points.
Recall Solution

Part A — (the honest minimum). . Setting gives the single critical point . Test either side:

  • : → decreasing.
  • : → increasing. Sign local minimum at . So genuinely has an extremum — it is not a counterexample to "zero slope ⇒ extremum," it just happens to be a case where the implication holds.

Part B — (the real trap). (chain rule: outer power , inner derivative ). Setting it to zero gives , one critical point. Since always (even power), everywhere:

  • : → increasing.
  • : → increasing. Sign same sign on both sides → no extremum. The point is an inflection with a horizontal tangent. Answer: has a local min at ; has a critical point at but no extremum. The difference: does change sign at , while (even power) does not.

Level 4 — Synthesis (combine ideas)

Recall Solution

Domain first. We need . So the domain is the closed interval . Step 1 — differentiate (product + chain rule). Write . Combine over one denominator: Step 2 — critical points.

  • : numerator zero → . Both lie inside . ✓
  • undefined: denominator zero at — but these are the endpoints of the domain, so we classify them separately with one-sided reasoning below. Step 3–4 — sign of (denominator on the interior, so the sign follows the numerator ):
  • : → decreasing on .
  • : → increasing on .
  • : → decreasing on . Step 5 — classify interior critical points.
  • At : local min. Value .
  • At : local max. Value . Step 6 — classify the endpoints (one-sided test). On a closed domain an endpoint is a local extremum if moves the "downhill" way as you step inward from it.
  • At the left endpoint : just to the right, on , (decreasing). So stepping inward from the function goes down, meaning sits above its immediate neighbours → local maximum. Value .
  • At the right endpoint : just to the left, on , , so approaching the function is still falling; that makes lower than its immediate neighbours → local minimum. Value . Answer: interior local min and local max ; endpoint local max and endpoint local min .
Recall Solution

Domain. is never zero → domain is all real . Step 1 — quotient rule. Step 2. Denominator always, so . No undefined points. Step 3–4 — sign follows numerator :

  • : → decreasing on .
  • : → increasing on .
  • : → decreasing on . Step 5.
  • At : local min, value .
  • At : local max, value . Answer: local min , local max .

Level 5 — Mastery (prove / generalise)

Recall Solution

Step 1. . Reasoning. A local extremum needs a critical point where changes sign. Since is an upward parabola in :

  • If : for all (never zero) → never zero, strictly increasing → no extrema. ✓
  • If : , zero only at but (no sign change) → no extrema (inflection). ✓
  • If : has two roots , and the parabola dips below zero between them → goes → genuine local max then local min → has extrema. Answer: has no local extrema exactly when .
Recall Solution

Goal. Show: for any in , (the definition of increasing). Setup. Since is differentiable on (that's what " exists" means), it is also continuous on and differentiable on — exactly the hypotheses of the Mean Value Theorem. Apply MVT. There exists some with Why MVT and not something else? We need to convert a statement about the slope into a statement about values vs . MVT is the exact bridge: it equates the average slope over to an actual instantaneous slope somewhere inside. Finish. Rearrange: . By hypothesis , and since . A positive times a positive is positive, so , i.e. . As were arbitrary, is increasing on .

Recall Solution

Step 1. . Step 2 — interior critical points. . Before dividing by , check the points where (namely ) are not lost:

  • at : .
  • at : . Neither is critical, so dividing was safe. On , gives and . Step 3–4 — sign of :
  • : → increasing on .
  • : → decreasing on .
  • : → increasing on . Step 5 — interior critical points.
  • At : local max, value .
  • At : local min, value . Step 6 — endpoints (one-sided test).
  • Left endpoint : just to the right, on , (increasing). Stepping inward the function rises, so is below its neighbours → local minimum, value .
  • Right endpoint : just to the left, on , (increasing), so approaching the function is still rising; that makes above its neighbours → local maximum, value . Answer: interior local max , interior local min ; endpoint local min and endpoint local max .

Connections

  • Mean Value Theorem — the engine we prove monotonicity with (L5·Q2).
  • Fermat's Theorem on Stationary Points — why we only search critical points.
  • Second Derivative Test — a faster classifier when is easy.
  • Concavity and Inflection Points — the inflection at lives here.
  • Optimization (Closed Interval Method) — extend L4·Q1's endpoints to global extrema.
  • Curve Sketching — assemble all these results into a graph.