Resources

A student seeks the absolute maximum and minimum of f(x,y)=xyf(x, y) = xy subject to the constraint g(x,y)=x2+4y2−8=0g(x, y) = x^2 + 4y^2 - 8 = 0. Compare Method 1 (Lagrange Multipliers: ∇f=λ∇g\nabla f = \lambda \nabla g) and Method 2 (Direct substitution of y2=8−x24y^2 = \frac{8 - x^2}{4} into f2(x,y)=x2y2f^2(x, y) = x^2 y^2). Explain the counter-intuitive trap that arises if Method 2 is applied without enforcing domain constraints on xx, and state all four constrained critical points (x,y)(x, y) along with the maximum and minimum values of ff.

Answer and steps

Final answer

Maximum value is 22 at (2,1)(2, 1) and (−2,−1)(-2, -1); minimum value is −2-2 at (2,−1)(2, -1) and (−2,1)(-2, 1). The counter-intuitive trap in Method 2 arises because maximizing F(x)=x2(8−x24)=2x2−x44F(x) = x^2 \left(\frac{8 - x^2}{4}\right) = 2x^2 - \frac{x^4}{4} as an unconstrained polynomial ignores the implicit boundary condition 8−x2≥0  ⟹  x∈[−22,22]8 - x^2 \ge 0 \implies x \in [-2\sqrt{2}, 2\sqrt{2}], and fails to distinguish signs since squaring ff loses the negative extrema without checking the original sign of xyxy.

Steps

Method 1: Lagrange Multipliers

To find the extrema of f(x,y)=xyf(x, y) = xy subject to the constraint g(x,y)=x2+4y2−8=0g(x, y) = x^2 + 4y^2 - 8 = 0, set up the Lagrange condition ∇f=λ∇g\nabla f = \lambda \nabla g:

∇f=⟨y,x⟩,∇g=⟨2x,8y⟩\nabla f = \langle y, x \rangle, \quad \nabla g = \langle 2x, 8y \rangle

Equating components gives: y=2λxy = 2\lambda x x=8λyx = 8\lambda y

Multiply the first equation by xx and the second by yy: xy=2λx2,xy=8λy2xy = 2\lambda x^2, \quad xy = 8\lambda y^2

Equating the two expressions: 2λx2=8λy2  ⟹  λ(x2−4y2)=02\lambda x^2 = 8\lambda y^2 \implies \lambda (x^2 - 4y^2) = 0

  • If λ=0\lambda = 0, then x=0x = 0 and y=0y = 0, but (0,0)(0, 0) does not satisfy x2+4y2=8x^2 + 4y^2 = 8.
  • Therefore, λ≠0\lambda \ne 0, which requires x2=4y2x^2 = 4y^2.

Substitute x2=4y2x^2 = 4y^2 into the constraint: 4y2+4y2=8  ⟹  8y2=8  ⟹  y=±14y^2 + 4y^2 = 8 \implies 8y^2 = 8 \implies y = \pm 1

Using x2=4y2x^2 = 4y^2, we obtain x2=4x^2 = 4, so x=±2x = \pm 2.

This yields four critical points: (2,1),(−2,−1),(2,−1),(−2,1)(2, 1), \quad (-2, -1), \quad (2, -1), \quad (-2, 1)

Evaluating f(x,y)=xyf(x, y) = xy at these points:

  • f(2,1)=2f(2, 1) = 2
  • f(−2,−1)=2f(-2, -1) = 2
  • f(2,−1)=−2f(2, -1) = -2
  • f(−2,1)=−2f(-2, 1) = -2

Method 2: Direct Substitution into the Squared Function

From the constraint, express y2y^2 in terms of x2x^2: y2=8−x24y^2 = \frac{8 - x^2}{4}

Define F(x)=[f(x,y)]2=x2y2F(x) = [f(x, y)]^2 = x^2 y^2: F(x)=x2(8−x24)=2x2−x44F(x) = x^2 \left(\frac{8 - x^2}{4}\right) = 2x^2 - \frac{x^4}{4}

Find critical points of F(x)F(x) by setting F′(x)=0F'(x) = 0: F′(x)=4x−x3=x(4−x2)=0F'(x) = 4x - x^3 = x(4 - x^2) = 0

This gives x=0x = 0 or x=±2x = \pm 2.

  • For x=±2x = \pm 2, y2=8−44=1  ⟹  y=±1y^2 = \frac{8 - 4}{4} = 1 \implies y = \pm 1, giving F(±2)=4F(\pm 2) = 4.
  • For x=0x = 0, y2=2  ⟹  y=±2y^2 = 2 \implies y = \pm \sqrt{2}, giving F(0)=0F(0) = 0.

Since F(x)=[f(x,y)]2=4F(x) = [f(x, y)]^2 = 4 at x=±2x = \pm 2, we have f(x,y)=±2f(x, y) = \pm 2.


Pitfalls and Counter-Intuitive Traps of Method 2

  1. Implicit Domain Restrictions: Because yy must be real, y2≥0y^2 \ge 0, which imposes: 8−x2≥0  ⟹  x∈[−22,22]8 - x^2 \ge 0 \implies x \in [-2\sqrt{2}, 2\sqrt{2}] Treating F(x)F(x) as an unconstrained quartic polynomial over (−∞,∞)(-\infty, \infty) ignores the physical boundary of the ellipse. Although the local maxima x=±2x = \pm 2 happen to lie strictly inside this interval, failing to establish the domain can lead to false conclusions about global behavior (e.g., assuming F→−∞F \to -\infty as x→±∞x \to \pm\infty, rather than F=0F = 0 at the endpoints x=±22x = \pm 2\sqrt{2}).
  2. Loss of Sign Information: Squaring f(x,y)f(x, y) maps both positive and negative values to non-negative numbers: [f(x,y)]2=4  ⟹  f(x,y)=2orf(x,y)=−2[f(x, y)]^2 = 4 \implies f(x, y) = 2 \quad \text{or} \quad f(x, y) = -2 Maximizing F(x)F(x) simultaneously captures both the absolute maximum (+2+2) and the absolute minimum (−2-2). Method 2 alone cannot determine which coordinate pairs correspond to the maximum and which to the minimum without testing the sign of the product xyxy at each individual point.

Conclusion

  • Constrained Critical Points: (2,1)(2, 1), (−2,−1)(-2, -1), (2,−1)(2, -1), and (−2,1)(-2, 1)
  • Absolute Maximum Value: 22, achieved at (2,1)(2, 1) and (−2,−1)(-2, -1)
  • Absolute Minimum Value: −2-2, achieved at (2,−1)(2, -1) and (−2,1)(-2, 1)

Typical errors

OverviewACEpath Teacher's Diagnosis

01 The student ignores the restriction that y² must be non-negative when substituting into f², leading them to evaluate unconstrained critical points without checking real domain boundaries.

Sample Student Work

Reviewed by ACEpath Teacher
Step-by-Step Review
Walk Through It with Your Teacher
Correct
  1. Step 1

    ∇f=(y,x),∇g=(2x,8y)  ⟹  y=2λx,x=8λy  ⟹  x=8λ(2λx)=16λ2x  ⟹  16λ2=1  ⟹  λ=±14\nabla f = (y, x), \quad \nabla g = (2x, 8y) \implies y = 2\lambda x, \quad x = 8\lambda y \implies x = 8\lambda(2\lambda x) = 16\lambda^2 x \implies 16\lambda^2 = 1 \implies \lambda = \pm \frac{1}{4}

  2. Step 2

    y=±x2  ⟹  x2+4(x2)2=8  ⟹  2x2=8  ⟹  x=±2  ⟹  y=±1  ⟹  Points: (2,1),(−2,−1),(2,−1),(−2,1)  ⟹  Maximum value =2, Minimum value =−2y = \pm \frac{x}{2} \implies x^2 + 4\left(\frac{x}{2}\right)^2 = 8 \implies 2x^2 = 8 \implies x = \pm 2 \implies y = \pm 1 \implies \text{Points: } (2, 1), (-2, -1), (2, -1), (-2, 1) \implies \text{Maximum value } = 2, \text{ Minimum value } = -2

  3. Step 3

    f2=x2y2=x2(8−x24)=2x2−x44  ⟹  d(f2)dx=4x−x3=x(4−x2)=0  ⟹  x=0 or x=±2f^2 = x^2 y^2 = x^2\left(\frac{8 - x^2}{4}\right) = 2x^2 - \frac{x^4}{4} \implies \frac{d(f^2)}{dx} = 4x - x^3 = x(4 - x^2) = 0 \implies x = 0 \text{ or } x = \pm 2

  4. Step 4

    Taking x→∞ gives f2→−∞, so there is no minimum value for f2. The trap is that f2 appears unbounded below because the condition x2≤8 was ignored.\text{Taking } x \to \infty \text{ gives } f^2 \to -\infty, \text{ so there is no minimum value for } f^2. \text{ The trap is that } f^2 \text{ appears unbounded below because the condition } x^2 \le 8 \text{ was ignored.}

WHAT YOU DID WELL
Compliments from Your Teacher

You correctly calculated the gradients and solved the Lagrange multiplier system.

You found all four constrained critical points along with the correct maximum and minimum values.

You correctly expressed f2f^2 as a single-variable quartic polynomial and located its critical points.

You clearly identified that ignoring x2≤8x^2 \le 8 causes f2f^2 to falsely appear unbounded below.

THE ERROR
Where Your ACEpath Teacher Spotted the Mistake
none
REASONING PATH
Your ACEpath teacher’s suggested approach
Implicit Domain Enforcement
  • When substituting an expression such as y2=g(x)y^2 = g(x) into an objective function, you must enforce the reality constraint g(x)≥0g(x) \ge 0.

  • Always write down the interval for the independent variable before analyzing end behavior or global extrema.

Sign Recovery After Squaring
  • Squaring an objective function ff maps negative values to positive values ([f]2≥0[f]^2 \ge 0), conflating absolute maxima and minima.

  • Test the original sign of f(x,y)f(x, y) at every critical point found through f2f^2.

CORRECTED PATH
ACEpath Teacher's Demonstration
Final answer

Maximum value is 22 at (2,1)(2, 1) and (−2,−1)(-2, -1); minimum value is −2-2 at (2,−1)(2, -1) and (−2,1)(-2, 1).

NEXT STEPS
Final Takeaways from Your Teacher
  • Explore how the second-derivative test or bordered Hessian confirms the local nature of constrained critical points.

  • Practice solving constrained optimization problems where the constraint boundary points themselves yield the global extrema.

OverviewACEpath Teacher's Diagnosis

02 The student concludes that maximizing f² only gives maximum values, failing to take the negative square root to identify the minimum of f.

Sample Student Work

Reviewed by ACEpath Teacher
Step-by-Step Review
Walk Through It with Your Teacher
Partially correct
  1. Step 1

    y=2λx,x=8λy  ⟹  x2=4y2  ⟹  x2+x2=8  ⟹  2x2=8  ⟹  x=±2, y=±1y = 2\lambda x,\quad x = 8\lambda y \implies x^2 = 4y^2 \implies x^2 + x^2 = 8 \implies 2x^2 = 8 \implies x = \pm 2,\ y = \pm 1

  2. Step 2

    Critical points: (2,1),(−2,−1),(2,−1),(−2,1)  ⟹  Maximum=2, Minimum=0\text{Critical points: } (2,1), (-2,-1), (2,-1), (-2,1) \implies \text{Maximum} = 2,\ \text{Minimum} = 0

    Correct

    Critical points: (2,1),(−2,−1),(2,−1),(−2,1)  ⟹  f(2,1)=f(−2,−1)=2, f(2,−1)=f(−2,1)=−2\text{Critical points: } (2,1), (-2,-1), (2,-1), (-2,1) \implies f(2,1)=f(-2,-1)=2,\ f(2,-1)=f(-2,1)=-2

    You correctly listed all four critical points, but evaluating f(x,y)=xyf(x, y) = xy at (2,−1)(2, -1) and (−2,1)(-2, 1) yields f=−2f = -2, so the absolute minimum is −2-2, not 00.

  3. Step 3

    H(x)=2x2−x44  ⟹  H′(x)=4x−x3=0  ⟹  x2=4  ⟹  H(2)=4, so f=4=2H(x) = 2x^2 - \frac{x^4}{4} \implies H'(x) = 4x - x^3 = 0 \implies x^2 = 4 \implies H(2) = 4,\ \text{so } f = \sqrt{4} = 2

  4. Step 4

    Since f2≥0, the minimum of f occurs at f2=0, giving minimum value 0The trap is assuming the minimum of f corresponds to the minimum of f2 rather than −H\text{Since } f^2 \ge 0\text{, the minimum of } f \text{ occurs at } f^2 = 0\text{, giving minimum value } 0 \newline \text{The trap is assuming the minimum of } f \text{ corresponds to the minimum of } f^2 \text{ rather than } -\sqrt{H}

    Correct

    Implicit domain: y2=8−x24≥0  ⟹  x∈[−22,22].Without this constraint, H(x)→−∞ as x→±∞, creating a false global minimum. Furthermore, H=4  ⟹  f=±2.\text{Implicit domain: } y^2 = \frac{8 - x^2}{4} \ge 0 \implies x \in [-2\sqrt{2}, 2\sqrt{2}]. \newline \text{Without this constraint, } H(x) \to -\infty \text{ as } x \to \pm\infty\text{, creating a false global minimum. Furthermore, } H = 4 \implies f = \pm 2.

    You concluded that the minimum of ff is 00 and identified the trap solely as forgetting −H-\sqrt{H}. The question specifically asked for the trap that arises when applying Method 2 without enforcing domain constraints on xx: because yy must be real, 8−x2≥08 - x^2 \ge 0, restricting xx to [−22,22][-2\sqrt{2}, 2\sqrt{2}]. Without this boundary restriction, an unconstrained quartic H(x)=2x2−x44H(x) = 2x^2 - \frac{x^4}{4} appears to decrease without bound to −∞-\infty as x→±∞x \to \pm\infty, which falsely suggests H(x)H(x) has no global minimum.

WHAT YOU DID WELL
Compliments from Your Teacher

You correctly established the Lagrange multiplier system and found all four critical points on the ellipse.

You correctly eliminated y2y^2 to construct the squared objective function H(x)=2x2−x44H(x) = 2x^2 - \frac{x^4}{4}.

You recognized that squaring a function loses sign information and that negative roots must be considered.

THE ERROR
Where Your ACEpath Teacher Spotted the Mistake
Missing Domain Constraint & Extrema Evaluation

You evaluated the minimum of xyxy at (2,−1)(2, -1) and (−2,1)(-2, 1) as 00 rather than multiplying the coordinates to get −2-2.

WHY IT HAPPENED
ACEpath Teacher's Diagnosis

When converting a constrained multivariable problem into an unconstrained single-variable problem by substitution, you focused on the algebraic stationary points without identifying the hidden domain boundary imposed by real-valued variables (y2≥0y^2 \ge 0), and you conflated f2=0f^2 = 0 with the minimum of ff rather than evaluating the original function f(x,y)f(x, y) at the critical points.

REASONING PATH
Your ACEpath teacher’s suggested approach
Verify Critical Point Values in the Original Function
  • Always evaluate f(x,y)f(x, y) directly at all candidate points before declaring absolute extrema.

  • Substitute each candidate pair into f(x,y)=xyf(x, y) = xy: (2)(−1)=−2(2)(-1) = -2 and (−2)(1)=−2(-2)(1) = -2.

Identify Implicit Substitution Domains
  • Whenever you substitute an expression for y2y^2, enforce y2≥0y^2 \ge 0 to bound the independent variable.

  • Write y2=8−x24≥0  ⟹  8−x2≥0  ⟹  −22≤x≤22y^2 = \frac{8 - x^2}{4} \ge 0 \implies 8 - x^2 \ge 0 \implies -2\sqrt{2} \le x \le 2\sqrt{2}.

CORRECTED PATH
ACEpath Teacher's Demonstration
  1. Evaluate f(x,y)=xyf(x, y) = xy at all four critical points: f(2,1)=2f(2, 1) = 2, f(−2,−1)=2f(-2, -1) = 2, f(2,−1)=−2f(2, -1) = -2, and f(−2,1)=−2f(-2, 1) = -2.

  2. Identify that the absolute maximum is 22 and the absolute minimum is −2-2.

  3. State the domain trap: Substituting y2=8−x24y^2 = \frac{8 - x^2}{4} requires y2≥0y^2 \ge 0, so x∈[−22,22]x \in [-2\sqrt{2}, 2\sqrt{2}]. If this is ignored, the polynomial H(x)=2x2−x44H(x) = 2x^2 - \frac{x^4}{4} diverges to −∞-\infty as x→±∞x \to \pm\infty, creating a misleading unconstrained picture.

Final answer

Maximum value is 22 at (2,1)(2, 1) and (−2,−1)(-2, -1); minimum value is −2-2 at (2,−1)(2, -1) and (−2,1)(-2, 1). The domain trap in Method 2 is ignoring 8−x2≥0  ⟹  x∈[−22,22]8 - x^2 \ge 0 \implies x \in [-2\sqrt{2}, 2\sqrt{2}], which hides the closed boundary of the ellipse and makes H(x)H(x) appear unbounded below.

NEXT STEPS
Final Takeaways from Your Teacher
  • Practice stating the real-variable domain constraint whenever eliminating squared variables like y2y^2 or sin⁡2(θ)\sin^2(\theta).

  • Check both positive and negative values of ff when maximizing f2f^2, and verify them directly against the original multivariable objective function.

Lock it in

Similar question

01

Find the absolute maximum and minimum of f(x,y)=x+2yf(x, y) = x + 2y subject to x2+4y2=8x^2 + 4y^2 = 8 using Lagrange multipliers.

Answer

Set ∇f=⟨1,2⟩=λ⟨2x,8y⟩\nabla f = \langle 1, 2 \rangle = \lambda \langle 2x, 8y \rangle. This gives 1=2λx1 = 2\lambda x and 2=8λy2 = 8\lambda y, so x=12λx = \frac{1}{2\lambda} and y=14λ=x2y = \frac{1}{4\lambda} = \frac{x}{2}. Substituting into the constraint yields x2+4(x/2)2=2x2=8  ⟹  x=±2x^2 + 4(x/2)^2 = 2x^2 = 8 \implies x = \pm 2 and y=±1y = \pm 1. The maximum value is f(2,1)=4f(2, 1) = 4 and the minimum value is f(−2,−1)=−4f(-2, -1) = -4.

02

A student attempts to find the extrema of f(x,y)=xy2f(x, y) = x y^2 subject to x2+y2=3x^2 + y^2 = 3 by substituting y2=3−x2y^2 = 3 - x^2 to get h(x)=x(3−x2)=3x−x3h(x) = x(3 - x^2) = 3x - x^3. State the required domain for xx and determine all critical points and absolute extrema of ff.

Answer

Because y2≥0y^2 \ge 0, we require 3−x2≥0  ⟹  x∈[−3,3]3 - x^2 \ge 0 \implies x \in [-\sqrt{3}, \sqrt{3}]. Setting h′(x)=3−3x2=0h'(x) = 3 - 3x^2 = 0 gives x=±1x = \pm 1. For x=1x = 1, y2=2  ⟹  y=±2y^2 = 2 \implies y = \pm\sqrt{2}, giving maximum value f(1,±2)=2f(1, \pm\sqrt{2}) = 2. For x=−1x = -1, y2=2  ⟹  y=±2y^2 = 2 \implies y = \pm\sqrt{2}, giving minimum value f(−1,±2)=−2f(-1, \pm\sqrt{2}) = -2. At boundary points x=±3x = \pm\sqrt{3}, y=0y = 0 and f=0f = 0.

03

Consider maximizing f(x,y)=x2−2y2f(x, y) = x^2 - 2y^2 subject to x2+y2=4x^2 + y^2 = 4. Compare what happens if you substitute x2=4−y2x^2 = 4 - y^2 to obtain k(y)=4−3y2k(y) = 4 - 3y^2 over its natural domain versus ignoring domain boundaries.

Answer

The condition x2=4−y2≥0x^2 = 4 - y^2 \ge 0 restricts y∈[−2,2]y \in [-2, 2]. On this interval, k(y)=4−3y2k(y) = 4 - 3y^2 has an interior maximum at y=0y = 0 where f=4f = 4 (at (±2,0)(\pm 2, 0)), and endpoint minima at y=±2y = \pm 2 where f=−8f = -8 (at (0,±2)(0, \pm 2)). If the domain is ignored, the parabola 4−3y24 - 3y^2 appears unbounded below as y→±∞y \to \pm\infty, missing the fact that the minimum is constrained to −8-8.

Loading workspace…