MATH 140 Calculus 1 • McGill University, Montreal

Corrected exercises: functions, graphs and transformations (MATH 140)

This is the corrected exercise set for the review of functions that opens MATH 140, Calculus 1, at McGill University, sections 1.1 to 1.3 of Stewart. Nothing here is new in principle, and that is the danger: every later chapter of the course differentiates, takes limits of, or sketches these functions, and a domain read on the wrong formula or a shift in the wrong direction costs marks again on the derivative, the asymptote and the curve sketch. Every number is exact and chosen to be done by hand, and each solution names the rule it applies.

The thread running through the whole set: read a formula from the INSIDE out. What happens inside the parentheses acts on xx and runs backwards, so f(x+3)f(x + 3) moves LEFT and f(2x+4)f(2x + 4) must be factored as f(2(x+2))f(2(x + 2)) before it is read. What happens outside acts on yy and runs as written, in the order of operations. The same reading gives the domain of a composition, inner function first, the parity test, −x-x inside and compare outside, and the value of a piecewise function, which piece first, then which formula.

The traps named in the solutions: reading the domain on a simplified formula, copying open dots into the range, giving the range of a quadratic from its two ends, a2=a\sqrt{a^2} = a instead of ∣a∣|a|, keeping a candidate that is not in its own piece, proving a symmetry with one point, forgetting that parity needs a symmetric domain, shifting the wrong way, compressing when the formula stretches, reading f(2x+4)f(2x + 4) as a shift of 44, adding before multiplying outside, confusing composition with a product, and extrapolating a model outside its domain.

10 corrected exercises • 100 points • 150 minutes

Revision sheet for this chapter → Every MATH 140 chapter →

Self-checking set Type your answers under each question: the page tells you right or wrong before you open the solution. With an account, every answer right on the first try earns points.

Course recap

  • • Domain in this chapter: a denominator is never 00, the quantity under an even root is ≥0\ge 0 (>0> 0 if the root is in a denominator); odd roots accept every real number.
  • • ∣a∣=a|a| = a if a≥0a \ge 0, −a-a if a<0a < 0, and a2=∣a∣\sqrt{a^2} = |a|. A piecewise equation is solved piece by piece, and each candidate is checked against its own interval.
  • • ff even: f(−x)=f(x)f(-x) = f(x), symmetric about the yy-axis. ff odd: f(−x)=−f(x)f(-x) = -f(x), symmetric about the origin. Both need a domain symmetric about 00.
  • • y=af(b(x−h))+ky = af(b(x - h)) + k: the point (u,w)(u, w) of ff goes to (ub+h, aw+k)\left(\frac{u}{b} + h,\ aw + k\right). Inside: backwards. Outside: as written, multiply before adding.
  • • (f∘g)(x)=f(g(x))(f \circ g)(x) = f(g(x)): xx in the domain of gg AND g(x)g(x) in the domain of ff. In general f∘g≠g∘ff \circ g \ne g \circ f and f∘g≠fgf \circ g \ne fg.
  • • Linear y=mx+by = mx + b: equal steps in xx add mm to yy. Power y=kxpy = kx^p: doubling xx multiplies yy by 2p2^p. A model has a domain of validity.

Part A: the basics (/50)

Exercise 1: Domain and range, from a graph and from a formula

The domain of a function is the set of inputs it accepts; its range is the set of outputs it actually produces. On a graph, the domain is read on the xx-axis and the range on the yy-axis; a filled dot is a point of the graph, an open dot is not. From a formula, a real-valued expression has only two restrictions in this chapter: a denominator cannot be 00, and the quantity under an EVEN root must be ≥0\ge 0. An odd root, such as  3\sqrt[3]{\ }, accepts every real number.

The figure shows the graph of a function ff made of two line segments.

-4-3-2-112345-2-112345y = f(x)x
  • a) Read the domain and the range of ff on the figure, and give f(1)f(1).
  • b) Find the domain of g(x)=x+3x2−4xg(x) = \frac{\sqrt{x + 3}}{x^2 - 4x}.
  • c) Find the domain of p(x)=x−83x2−5x+4p(x) = \frac{\sqrt[3]{x - 8}}{\sqrt{x^2 - 5x + 4}}.
  • d) Find the range of q(x)=x2+4x+1q(x) = x^2 + 4x + 1 on the interval [−3,1][-3, 1], without a derivative.
  • e) Are F(x)=x2−9x−3F(x) = \frac{x^2 - 9}{x - 3} and G(x)=x+3G(x) = x + 3 the same function? Describe the graph of FF and give its range.

Type your answers, the page tells you right or wrong 0/11

a)
Domain of ff ,
Range of ff ,
b)
c)
d)
Range of qq on [−3,1][-3, 1] ,
e)
Show the solution

Answers

  • a) Domain [−3,4)[-3, 4), range [−1,4][-1, 4], f(1)=4f(1) = 4
  • b) [−3,0)∪(0,4)∪(4,∞)[-3, 0) \cup (0, 4) \cup (4, \infty)
  • c) (−∞,1)∪(4,∞)(-\infty, 1) \cup (4, \infty)
  • d) q(x)=(x+2)2−3q(x) = (x + 2)^2 - 3: range [−3,6][-3, 6]
  • e) No: 33 is not in the domain of FF. The graph of FF is the line y=x+3y = x + 3 with a hole at (3,6)(3, 6); range R∖{6}\mathbb{R} \setminus \{6\}.

a) Domain: the graph starts with a FILLED dot at x=−3x = -3 and ends with an OPEN dot at x=4x = 4, and there is no gap in between, since the second segment starts at x=1x = 1 exactly where the first one stops. So the domain is [−3,4)[-3, 4). Value at 11: above x=1x = 1 there are two dots, (1,3)(1, 3) open and (1,4)(1, 4) filled; only the filled one is on the graph, so f(1)=4f(1) = 4. Range: the first segment, y=x+2y = x + 2 for −3≤x<1-3 \le x < 1, produces every value of [−1,3)[-1, 3); the second, y=5−xy = 5 - x for 1≤x<41 \le x < 4, produces every value of (1,4](1, 4]. The range is the union, [−1,3)∪(1,4]=[−1,4][-1, 3) \cup (1, 4] = [-1, 4]. The trap is to copy the open dots into the range: 33 is missed at the open dot (1,3)(1, 3) but attained on the second segment at x=2x = 2, where 5−2=35 - 2 = 3; and 11, missed at (4,1)(4, 1), is attained at x=−1x = -1 on the first segment. An open dot removes a POINT of the graph, not automatically a value of the range.

b) Two restrictions, taken one at a time. The even root needs x+3≥0x + 3 \ge 0, so x≥−3x \ge -3. The denominator needs x2−4x=x(x−4)≠0x^2 - 4x = x(x - 4) \ne 0, so x≠0x \ne 0 and x≠4x \ne 4. Both must hold at once: the domain is [−3,∞)[-3, \infty) with 00 and 44 removed, that is [−3,0)∪(0,4)∪(4,∞)[-3, 0) \cup (0, 4) \cup (4, \infty). Check with a test value: g(−3)=021=0g(-3) = \frac{0}{21} = 0 exists, so −3-3 belongs to the domain and the bracket is closed there. Forgetting to factor the denominator and removing only x=4x = 4 costs the mark.

c) The cube root accepts every real number, including negative ones, so the numerator imposes NOTHING: −83=−2\sqrt[3]{-8} = -2 is a real number. The denominator is an even root, and it is a denominator, so what is under it must be STRICTLY positive: x2−5x+4=(x−1)(x−4)>0x^2 - 5x + 4 = (x - 1)(x - 4) > 0. A product of two factors is positive when both have the same sign: x<1x < 1 or x>4x > 4. The domain is (−∞,1)∪(4,∞)(-\infty, 1) \cup (4, \infty). Two classic errors: writing ≥0\ge 0, which would accept x=1x = 1 and divide by 00; and requiring x−8≥0x - 8 \ge 0 as if the cube root were a square root, which would give the wrong domain [8,∞)[8, \infty).

d) Complete the square: x2+4x+1=(x2+4x+4)−3=(x+2)2−3x^2 + 4x + 1 = (x^2 + 4x + 4) - 3 = (x + 2)^2 - 3. On [−3,1][-3, 1], the quantity x+2x + 2 runs over [−1,3][-1, 3], so its square runs over [0,9][0, 9]: it reaches 00 at x=−2x = -2, which IS in the interval, and its largest value 99 at x=1x = 1. Hence qq runs over [0−3,9−3]=[−3,6][0 - 3, 9 - 3] = [-3, 6]. The trap is to evaluate the two ends only, q(−3)=−2q(-3) = -2 and q(1)=6q(1) = 6, and answer [−2,6][-2, 6]: the vertex (−2,−3)(-2, -3) lies inside the interval, lower than both ends. The ends give the range only when the function moves in one direction over the whole interval.

e) No. Two functions are equal when they have the same domain AND the same values. FF is not defined at x=3x = 3 (denominator 00), while G(3)=6G(3) = 6. Everywhere else, x2−9=(x−3)(x+3)x^2 - 9 = (x - 3)(x + 3) and the factor x−3≠0x - 3 \ne 0 can be cancelled: F(x)=x+3F(x) = x + 3 for x≠3x \ne 3. So the graph of FF is the line y=x+3y = x + 3 with ONE point removed, an open dot at (3,6)(3, 6). The only value the line would take there is 66, and no other xx gives x+3=6x + 3 = 6, so the range of FF is R∖{6}\mathbb{R} \setminus \{6\}. Simplifying first and reading the domain on the simplified formula is the error the whole chapter warns against: the domain is read on the formula AS GIVEN.

Tick the exercises you have done or want to review: a free account, no password, keeps your ticks from one visit to the next and tells you which chapter to tackle next. Create your space, an email is enough.

Exercise 2: Piecewise functions and absolute value: which piece, then which value

A piecewise function uses DIFFERENT formulas on different intervals. To evaluate it, the first question is never the formula: it is which interval the input belongs to. To solve an equation, each piece is solved on its own, and every candidate must then be checked against the interval of ITS piece. The absolute value is the first example: ∣a∣=a|a| = a if a≥0a \ge 0 and ∣a∣=−a|a| = -a if a<0a < 0, and a2=∣a∣\sqrt{a^2} = |a| for every real aa.

The figure shows the graph of a function hh made of a line, an arc of the parabola y=x2y = x^2 and a horizontal ray.

-4-3-2-112345-2-112345y = h(x)x
  • a) Read h(−3)h(-3), h(−1)h(-1), h(2)h(2) and h(4)h(4) on the figure, then write h(x)h(x) as a piecewise formula.
  • b) Write f(x)=∣2x−6∣f(x) = |2x - 6| and k(x)=∣x2−4∣k(x) = |x^2 - 4| without absolute value signs.
  • c) Solve ∣x−1∣=2x+4|x - 1| = 2x + 4.
  • d) A student simplifies x2−6x+9=x−3\sqrt{x^2 - 6x + 9} = x - 3. Correct the simplification for x<3x < 3 and check it at x=0x = 0.
  • e) Solve h(x)=2h(x) = 2.

Type your answers, the page tells you right or wrong 0/11

a)
b)
k(x)=4−x2k(x) = 4 - x^2 on the interval ,
c)
d)
e)
Show the solution

Answers

  • a) h(−3)=0h(-3) = 0, h(−1)=1h(-1) = 1, h(2)=4h(2) = 4, h(4)=1h(4) = 1; h(x)=x+3h(x) = x + 3 if x<−1x < -1, x2x^2 if −1≤x≤2-1 \le x \le 2, 11 if x>2x > 2
  • b) f(x)=2x−6f(x) = 2x - 6 if x≥3x \ge 3, 6−2x6 - 2x if x<3x < 3; k(x)=x2−4k(x) = x^2 - 4 if ∣x∣≥2|x| \ge 2, 4−x24 - x^2 if −2<x<2-2 < x < 2
  • c) x=−1x = -1 (x=−5x = -5 rejected)
  • d) x2−6x+9=∣x−3∣=3−x\sqrt{x^2 - 6x + 9} = |x - 3| = 3 - x for x<3x < 3; at x=0x = 0: 9=3\sqrt 9 = 3
  • e) x=2x = \sqrt 2 only (x=−1x = -1 and x=−2x = -\sqrt 2 rejected)

a) h(−3)=0h(-3) = 0: the line passes through (−3,0)(-3, 0). At x=−1x = -1 the figure shows an open dot at (−1,2)(-1, 2) and a filled dot at (−1,1)(-1, 1); the filled one is the graph, so h(−1)=1h(-1) = 1. At x=2x = 2 the filled dot is (2,4)(2, 4) and the open one (2,1)(2, 1), so h(2)=4h(2) = 4; and h(4)=1h(4) = 1 on the ray. Formula, piece by piece. The line goes through (−3,0)(-3, 0) and (−2,1)(-2, 1): slope 11, so y=x+3y = x + 3, and it stops at x=−1x = -1 with an open dot, so the interval is x<−1x < -1. The arc passes through (0,0)(0, 0), (1,1)(1, 1) and (2,4)(2, 4): y=x2y = x^2 on −1≤x≤2-1 \le x \le 2, both ends filled. The ray is y=1y = 1 for x>2x > 2. The inequalities, strict or not, ARE the open and filled dots, and they are part of the answer: each input must belong to exactly one piece.

b) 2x−6≥02x - 6 \ge 0 exactly when x≥3x \ge 3, so f(x)=2x−6f(x) = 2x - 6 for x≥3x \ge 3 and f(x)=−(2x−6)=6−2xf(x) = -(2x - 6) = 6 - 2x for x<3x < 3. For kk, the sign of x2−4=(x−2)(x+2)x^2 - 4 = (x - 2)(x + 2) decides: it is ≥0\ge 0 when x≤−2x \le -2 or x≥2x \ge 2, and negative when −2<x<2-2 < x < 2. So k(x)=x2−4k(x) = x^2 - 4 if x≤−2x \le -2 or x≥2x \ge 2, and k(x)=4−x2k(x) = 4 - x^2 if −2<x<2-2 < x < 2. Graphically, the part of the parabola below the xx-axis is reflected upward, and nothing else moves. The sign to study is the sign of what is INSIDE the bars, never the sign of xx.

c) Split on the sign of x−1x - 1. Case x≥1x \ge 1: x−1=2x+4x - 1 = 2x + 4 gives x=−5x = -5, which is NOT ≥1\ge 1: rejected. Case x<1x < 1: −(x−1)=2x+4-(x - 1) = 2x + 4 gives −3x=3-3x = 3, x=−1x = -1, which is <1< 1: kept. Check: ∣−1−1∣=2|-1 - 1| = 2 and 2(−1)+4=22(-1) + 4 = 2. A faster filter was available: an absolute value is never negative, so the right side needs 2x+4≥02x + 4 \ge 0, that is x≥−2x \ge -2, which eliminates −5-5 at once, since ∣−6∣=6|-6| = 6 while 2(−5)+4=−62(-5) + 4 = -6. Keeping x=−5x = -5 is the classic lost mark: the candidate solved its equation but not its case.

d) x2−6x+9=(x−3)2x^2 - 6x + 9 = (x - 3)^2, so x2−6x+9=(x−3)2=∣x−3∣\sqrt{x^2 - 6x + 9} = \sqrt{(x - 3)^2} = |x - 3|, NOT x−3x - 3. For x<3x < 3, x−3x - 3 is negative, so ∣x−3∣=3−x|x - 3| = 3 - x. Check at x=0x = 0: 9=3\sqrt{9} = 3, and 3−0=33 - 0 = 3, while the student's x−3x - 3 gives −3-3, a negative square root, impossible. The rule a2=a\sqrt{a^2} = a holds only for a≥0a \ge 0; in general a2=∣a∣\sqrt{a^2} = |a|. This is the inside-outside reading again: the square root sees (x−3)2(x - 3)^2, which has forgotten the sign of x−3x - 3.

e) Solve on each piece, then check each candidate against its interval. First piece: x+3=2x + 3 = 2 gives x=−1x = -1, but the first piece is x<−1x < -1: rejected (and indeed h(−1)=1h(-1) = 1, not 22; the point (−1,2)(-1, 2) is the open dot). Second piece: x2=2x^2 = 2 gives x=2x = \sqrt 2 or x=−2x = -\sqrt 2; the piece is −1≤x≤2-1 \le x \le 2, and 2≈1.41\sqrt 2 \approx 1.41 is in it, while −2≈−1.41-\sqrt 2 \approx -1.41 is not: only 2\sqrt 2 is kept. Third piece: 1=21 = 2 has no solution. Answer: x=2x = \sqrt 2, and only that. On the figure, the horizontal line y=2y = 2 meets the graph once, on the arc; it only touches the open dot at (−1,2)(-1, 2), which is not a point of the graph.

Exercise 3: Even, odd or neither: decided by computing f(-x)

A function ff whose domain is symmetric about 00 is EVEN if f(−x)=f(x)f(-x) = f(x) for every xx of its domain (graph symmetric about the yy-axis), and ODD if f(−x)=−f(x)f(-x) = -f(x) for every xx (graph symmetric about the origin). The test is a computation: replace xx by −x-x INSIDE, simplify, and compare the result with f(x)f(x) and with −f(x)-f(x). One numerical example can DISPROVE a symmetry; it can never prove one.

The figure shows the graph of a function ff defined on [−4,4][-4, 4], drawn for 0≤x≤40 \le x \le 4 only.

-5-4-3-2-112345-3-2-1123y = f(x)to completex
  • a) Decide whether each function is even, odd or neither: f1(x)=x4−3x2+1f_1(x) = x^4 - 3x^2 + 1, f2(x)=x3−4xf_2(x) = x^3 - 4x, f3(x)=x3+x2f_3(x) = x^3 + x^2.
  • b) Same question for k(x)=xx2+1k(x) = \frac{x}{x^2 + 1}, m(x)=(x+1)2m(x) = (x + 1)^2 and n(x)=x∣x∣n(x) = x|x|.
  • c) Complete the graph of the figure on [−4,0][-4, 0] if ff is even, then if ff is odd. Why must an odd function defined at 00 satisfy f(0)=0f(0) = 0?
  • d) A student notes that p(x)=x3−xp(x) = x^3 - x satisfies p(−1)=p(1)p(-1) = p(1) and concludes that pp is even. Correct the conclusion.
  • e) Is s(x)=x2s(x) = x^2, defined on [−1,3][-1, 3], even? Then prove that the product of an even function and an odd function is odd.

Type your answers, the page tells you right or wrong 0/15

a)
b)
c)
d)
e)
Show the solution

Answers

  • a) f1f_1 even, f2f_2 odd, f3f_3 neither
  • b) kk odd, mm neither, nn odd
  • c) Even: (−1,2)(-1, 2), (−3,−1)(-3, -1), (−4,0)(-4, 0); odd: (−1,−2)(-1, -2), (−3,1)(-3, 1), (−4,0)(-4, 0). f(0)=−f(0)f(0) = -f(0) forces f(0)=0f(0) = 0.
  • d) p(−2)=−6≠p(2)=6p(-2) = -6 \ne p(2) = 6: pp is not even; p(−x)=−p(x)p(-x) = -p(x), so pp is odd.
  • e) Neither: the domain is not symmetric. (EO)(−x)=E(x)⋅(−O(x))=−(EO)(x)(EO)(-x) = E(x)\cdot(-O(x)) = -(EO)(x).

a) f1(−x)=(−x)4−3(−x)2+1=x4−3x2+1=f1(x)f_1(-x) = (-x)^4 - 3(-x)^2 + 1 = x^4 - 3x^2 + 1 = f_1(x): even, every power of xx is even. f2(−x)=(−x)3−4(−x)=−x3+4x=−(x3−4x)=−f2(x)f_2(-x) = (-x)^3 - 4(-x) = -x^3 + 4x = -(x^3 - 4x) = -f_2(x): odd. f3(−x)=−x3+x2f_3(-x) = -x^3 + x^2, which is neither x3+x2x^3 + x^2 nor −x3−x2-x^3 - x^2. To be sure it is NEITHER, one counterexample for each symmetry is enough: f3(1)=2f_3(1) = 2 and f3(−1)=0f_3(-1) = 0, and 0≠20 \ne 2 (not even), 0≠−20 \ne -2 (not odd). A sum of an even part and an odd part, both nonzero, is neither. The two domains are R\mathbb{R}, symmetric, as the test requires.

b) k(−x)=−x(−x)2+1=−xx2+1=−k(x)k(-x) = \frac{-x}{(-x)^2 + 1} = -\frac{x}{x^2 + 1} = -k(x): odd. For mm, the look is misleading: it IS a square, but a square of x+1x + 1, not of xx. m(1)=4m(1) = 4 and m(−1)=0m(-1) = 0, so mm is neither even nor odd; its graph is symmetric about the line x=−1x = -1, not about the yy-axis. For nn: n(−x)=(−x)∣−x∣=−x∣x∣=−n(x)n(-x) = (-x)|-x| = -x|x| = -n(x), since ∣−x∣=∣x∣|-x| = |x|: odd. The minus sign inside the absolute value disappears, the one outside stays.

c) Even: each point (a,b)(a, b) of the graph has its mirror (−a,b)(-a, b). The vertices (1,2)(1, 2), (3,−1)(3, -1), (4,0)(4, 0) give (−1,2)(-1, 2), (−3,−1)(-3, -1), (−4,0)(-4, 0), joined by segments. Odd: each point (a,b)(a, b) has its image (−a,−b)(-a, -b), a half-turn about the origin: (−1,−2)(-1, -2), (−3,1)(-3, 1), (−4,0)(-4, 0). The solution figure shows both completions. For the last question, take x=0x = 0 in f(−x)=−f(x)f(-x) = -f(x): f(0)=−f(0)f(0) = -f(0), so 2f(0)=02f(0) = 0 and f(0)=0f(0) = 0. The drawn graph passes through the origin, so the odd completion is possible; if it had started at (0,1)(0, 1), no odd function could extend it.

d) A single point proves nothing about a statement that must hold for EVERY xx. Here p(1)=0p(1) = 0 and p(−1)=0p(-1) = 0 by coincidence, because ±1\pm 1 are roots. At x=2x = 2: p(2)=6p(2) = 6 and p(−2)=−8+2=−6≠6p(-2) = -8 + 2 = -6 \ne 6, so pp is NOT even. The algebra decides: p(−x)=−x3+x=−(x3−x)=−p(x)p(-x) = -x^3 + x = -(x^3 - x) = -p(x) for every xx, so pp is odd. Since pp is odd and p(1)=0p(1) = 0, also p(−1)=−0=0p(-1) = -0 = 0: the coincidence of the student is exactly what oddness predicts at a root.

e) Parity requires a domain symmetric about 00: if xx is in the domain, −x-x must be too. Here 22 is in [−1,3][-1, 3] but −2-2 is not, so s(−2)s(-2) does not exist and ss is neither even nor odd, although its formula is x2x^2. Parity belongs to the function, formula AND domain. For the product, let EE be even and OO odd with the same symmetric domain: (EO)(−x)=E(−x) O(−x)=E(x) (−O(x))=−E(x)O(x)=−(EO)(x)(EO)(-x) = E(-x)\,O(-x) = E(x)\,(-O(x)) = -E(x)O(x) = -(EO)(x), so EOEO is odd. Check with E(x)=x2E(x) = x^2, O(x)=xO(x) = x: x3x^3 is odd. The signs multiply like +1+1 and −1-1: even times odd is odd, odd times odd is even.

-5-4-3-2-112345-3-2-1123even: (-a, b)odd: (-a, -b)y = f(x)x

Exercise 4: Transformations of a graph, point by point and in the right order

Every transformation of y=f(x)y = f(x) acts on ONE variable. A change INSIDE the parentheses acts on xx and runs backwards: f(x−3)f(x - 3) moves the graph 33 units to the RIGHT, f(2x)f(2x) COMPRESSES it horizontally by a factor 22. A change OUTSIDE acts on yy and runs as written: f(x)+1f(x) + 1 moves it up 11, −2f(x)-2f(x) stretches it vertically by 22 and reflects it in the xx-axis. The safest method for a point: if (a,b)(a, b) is on the graph of ff, find the xx that makes the inside equal to aa, then apply the outside operations to bb.

The figure shows the graph of ff, made of three segments through A(−2,0)A(-2, 0), B(0,4)B(0, 4), C(2,2)C(2, 2) and D(4,2)D(4, 2). Its domain is [−2,4][-2, 4] and its range is [0,4][0, 4].

-3-2-112345-112345ABCDy = f(x)x
  • a) Give the images of AA, BB, CC, DD on the graph of y=f(x−3)+1y = f(x - 3) + 1, and its domain and range.
  • b) Compare y=−f(x)+2y = -f(x) + 2 and y=−(f(x)+2)y = -(f(x) + 2): image of BB and range of each.
  • c) Give the images of the four points on the graph of y=f(2x+4)y = f(2x + 4), and its domain. Describe the transformations in a correct order.
  • d) Give the images of BB and CC on the graphs of y=f(−x)y = f(-x) and y=−f(−x)y = -f(-x). What single geometric transformation is y=−f(−x)y = -f(-x)?
  • e) Find the domain and the range of y=1−2f(x2)y = 1 - 2f\left(\frac{x}{2}\right) without drawing it.

Type your answers, the page tells you right or wrong 0/17

a)
Domain ,
Range ,
b)
Range of −f(x)+2-f(x) + 2 ,
Range of −(f(x)+2)-(f(x) + 2) ,
c)
Domain ,
d)
e)
Domain ,
Range ,
Show the solution

Answers

  • a) (1,1)(1, 1), (3,5)(3, 5), (5,3)(5, 3), (7,3)(7, 3); domain [1,7][1, 7], range [1,5][1, 5]
  • b) B↦(0,−2)B \mapsto (0, -2), range [−2,2][-2, 2]; B↦(0,−6)B \mapsto (0, -6), range [−6,−2][-6, -2]
  • c) (−3,0)(-3, 0), (−2,4)(-2, 4), (−1,2)(-1, 2), (0,2)(0, 2); domain [−3,0][-3, 0]; compress by 22, then left 22
  • d) f(−x)f(-x): B↦(0,4)B \mapsto (0, 4), C↦(−2,2)C \mapsto (-2, 2); −f(−x)-f(-x): B↦(0,−4)B \mapsto (0, -4), C↦(−2,−2)C \mapsto (-2, -2); a half-turn about the origin
  • e) Domain [−4,8][-4, 8], range [−7,1][-7, 1]

a) Inside, x−3=ax - 3 = a gives x=a+3x = a + 3: each point moves 33 units RIGHT. Outside, +1+1: each yy goes up 11. So (a,b)↦(a+3,b+1)(a, b) \mapsto (a + 3, b + 1): A(−2,0)↦(1,1)A(-2, 0) \mapsto (1, 1), B(0,4)↦(3,5)B(0, 4) \mapsto (3, 5), C(2,2)↦(5,3)C(2, 2) \mapsto (5, 3), D(4,2)↦(7,3)D(4, 2) \mapsto (7, 3). The domain moves with the xx-coordinates: [−2+3,4+3]=[1,7][-2 + 3, 4 + 3] = [1, 7]; the range with the yy-coordinates: [0+1,4+1]=[1,5][0 + 1, 4 + 1] = [1, 5]. Check one point in the formula: at x=3x = 3, f(3−3)+1=f(0)+1=5f(3 - 3) + 1 = f(0) + 1 = 5, as found. Moving LEFT for a minus sign is the error this check catches.

b) Both act only on yy, but in a different order. y=−f(x)+2y = -f(x) + 2: first the minus sign, then +2+2, so b↦−b+2b \mapsto -b + 2; BB goes to (0,−4+2)=(0,−2)(0, -4 + 2) = (0, -2), and the range [0,4][0, 4] becomes [−4+2,0+2]=[−2,2][-4 + 2, 0 + 2] = [-2, 2]. y=−(f(x)+2)=−f(x)−2y = -(f(x) + 2) = -f(x) - 2: first +2+2, then the minus sign, so b↦−(b+2)b \mapsto -(b + 2); BB goes to (0,−6)(0, -6) and the range becomes [−6,−2][-6, -2]. Outside the parentheses, the order is the ORDER OF OPERATIONS of the formula, exactly as when evaluating a number: the parentheses of the second formula force the addition first. Two graphs 44 units apart for the same three symbols in a different order.

c) Inside, solve 2x+4=a2x + 4 = a: x=a−42x = \frac{a - 4}{2}. So (a,b)↦(a−42,b)(a, b) \mapsto \left(\frac{a - 4}{2}, b\right): A↦(−3,0)A \mapsto (-3, 0), B↦(−2,4)B \mapsto (-2, 4), C↦(−1,2)C \mapsto (-1, 2), D↦(0,2)D \mapsto (0, 2), and the domain is [−3,0][-3, 0], obtained also by solving −2≤2x+4≤4-2 \le 2x + 4 \le 4. To describe it in words, FACTOR the inside first: f(2x+4)=f(2(x+2))f(2x + 4) = f(2(x + 2)) is a horizontal compression by 22 followed by a shift 22 units LEFT; check on AA: −2↦−1↦−3-2 \mapsto -1 \mapsto -3. The other correct order is shift left 44, then compress: −2↦−6↦−3-2 \mapsto -6 \mapsto -3. The WRONG reading is compress by 22 then shift left 44, which sends AA to −1−4=−5-1 - 4 = -5 and gives the domain [−5,−2][-5, -2]: a whole graph in the wrong place. The solution figure shows ff and the correct result.

d) y=f(−x)y = f(-x): inside, −x=a-x = a gives x=−ax = -a, so (a,b)↦(−a,b)(a, b) \mapsto (-a, b), a reflection in the yy-axis: B(0,4)↦(0,4)B(0, 4) \mapsto (0, 4), which stays because it is ON the axis, and C(2,2)↦(−2,2)C(2, 2) \mapsto (-2, 2). y=−f(−x)y = -f(-x): the same inside, then a minus sign outside, (a,b)↦(−a,−b)(a, b) \mapsto (-a, -b): B↦(0,−4)B \mapsto (0, -4) and C↦(−2,−2)C \mapsto (-2, -2). Changing the signs of both coordinates is a half-turn (rotation by 180∘180^\circ) about the origin. This is also why a function is odd exactly when −f(−x)=f(x)-f(-x) = f(x): its graph is unchanged by the half-turn.

e) Domain: the inside x2\frac{x}{2} must be in the domain of ff: −2≤x2≤4-2 \le \frac{x}{2} \le 4, so −4≤x≤8-4 \le x \le 8. The domain is [−4,8][-4, 8], a horizontal STRETCH by 22. Range: the inside only chooses WHICH values of ff are used, and as xx runs over [−4,8][-4, 8], f(x2)f\left(\frac{x}{2}\right) takes every value of [0,4][0, 4]. Then outside, in the order of the formula: times −2-2 gives [−8,0][-8, 0] (the ends swap when multiplying by a negative number), then 1+1 + gives [−7,1][-7, 1]. Check with BB: x=0x = 0 gives 1−2f(0)=1−8=−71 - 2f(0) = 1 - 8 = -7, the lowest value. Writing [1−0,1−8]=[1,−7][1 - 0, 1 - 8] = [1, -7], with the ends in the wrong order, is the typical slip.

-4-3-2-112345-112345y = f(x)y = f(2x + 4)x

Exercise 5: From a basic graph to a formula: completing the square and reading the transformations

Most functions of the course are transformations of a few basic graphs: y=x2y = x^2, y=xy = \sqrt x, y=1xy = \frac{1}{x}, y=∣x∣y = |x|. To see which transformations, rewrite the formula so that the inside is a multiple of (x−h)(x - h) and the outside is a⋅( )+ka \cdot (\ ) + k: completing the square for a quadratic, factoring the coefficient of xx under a root, dividing for a quotient of two linear expressions. Then read the inside backwards and the outside as written.

The figure shows the graph of a function gg with a corner at (−1,3)(-1, 3), passing through (−3,−1)(-3, -1) and (1,−1)(1, -1).

-4-3-2-1123-2-11234y = g(x)x
  • a) Write y=x2−6x+5y = x^2 - 6x + 5 as y=(x−h)2+ky = (x - h)^2 + k. Describe the transformations from y=x2y = x^2, and give the vertex, the xx-intercepts and the range.
  • b) Same questions for y=−2x2−8x−5y = -2x^2 - 8x - 5, stating the order of the vertical operations.
  • c) Describe how the graph of y=8−2xy = \sqrt{8 - 2x} is obtained from y=xy = \sqrt x. Give its domain and check one point.
  • d) Show that x+1x−1=1+2x−1\frac{x + 1}{x - 1} = 1 + \frac{2}{x - 1} and describe the graph as a transformation of y=1xy = \frac{1}{x}. Give the domain and the range.
  • e) Find a formula g(x)=a∣x−h∣+kg(x) = a|x - h| + k for the function of the figure, and its xx-intercepts.

Type your answers, the page tells you right or wrong 0/18

a)
Range ,
b)
Range ,
c)
Domain ,
d)
e)
Show the solution

Answers

  • a) (x−3)2−4(x - 3)^2 - 4: right 33, down 44; vertex (3,−4)(3, -4); intercepts 11 and 55; range [−4,∞)[-4, \infty)
  • b) −2(x+2)2+3-2(x + 2)^2 + 3: left 22, stretch by 22, reflect, then up 33; vertex (−2,3)(-2, 3); range (−∞,3](-\infty, 3]
  • c) −2(x−4)\sqrt{-2(x - 4)}: compress by 22, reflect in the yy-axis, right 44; domain (−∞,4](-\infty, 4]; (2,2)(2, 2) is on it
  • d) Stretch by 22, right 11, up 11; domain x≠1x \ne 1, range y≠1y \ne 1
  • e) g(x)=−2∣x+1∣+3g(x) = -2|x + 1| + 3; x=12x = \frac{1}{2} and x=−52x = -\frac{5}{2}

a) Half of −6-6 is −3-3: x2−6x+5=(x2−6x+9)−9+5=(x−3)2−4x^2 - 6x + 5 = (x^2 - 6x + 9) - 9 + 5 = (x - 3)^2 - 4. Inside, x−3x - 3: shift 33 units RIGHT; outside, −4-4: shift 44 units down. The vertex (0,0)(0, 0) of y=x2y = x^2 goes to (3,−4)(3, -4). Intercepts: (x−3)2=4(x - 3)^2 = 4 gives x−3=±2x - 3 = \pm 2, so x=1x = 1 or x=5x = 5; check 1−6+5=01 - 6 + 5 = 0. Since (x−3)2≥0(x - 3)^2 \ge 0, y≥−4y \ge -4 with equality at x=3x = 3: the range is [−4,∞)[-4, \infty). Adding 99 without subtracting it again is the arithmetic slip of this step; checking the vertex value 32−18+5=−43^2 - 18 + 5 = -4 catches it.

b) Factor −2-2 from the terms in xx only: −2x2−8x−5=−2(x2+4x)−5=−2[(x+2)2−4]−5=−2(x+2)2+8−5=−2(x+2)2+3-2x^2 - 8x - 5 = -2(x^2 + 4x) - 5 = -2\left[(x + 2)^2 - 4\right] - 5 = -2(x + 2)^2 + 8 - 5 = -2(x + 2)^2 + 3. Inside, x+2x + 2: shift LEFT 22. Outside, the formula says: multiply by −2-2, THEN add 33. So stretch vertically by 22 and reflect in the xx-axis, and only then shift up 33. Shifting up 33 first would give −2[(x+2)2+3]=−2(x+2)2−6-2\left[(x + 2)^2 + 3\right] = -2(x + 2)^2 - 6, a different parabola. Vertex (−2,3)(-2, 3); since −2(x+2)2≤0-2(x + 2)^2 \le 0, the range is (−∞,3](-\infty, 3]. Forgetting to multiply the −4-4 by −2-2 when leaving the bracket gives +3−8+3 - 8 instead of +8−5+8 - 5, the most common error here.

c) Factor the coefficient of xx inside the root: 8−2x=−2(x−4)8 - 2x = -2(x - 4), so y=−2(x−4)y = \sqrt{-2(x - 4)}. Reading the inside backwards: x↦2xx \mapsto 2x compresses horizontally by 22, the minus sign reflects in the yy-axis, and x−4x - 4 shifts 44 units RIGHT, applied in that order to the graph of x\sqrt x: (a,b)↦(4−a2,b)(a, b) \mapsto \left(4 - \frac{a}{2}, b\right). The starting point (0,0)(0, 0) goes to (4,0)(4, 0) and (4,2)(4, 2) goes to (2,2)(2, 2). Check: 8−2⋅2=4=2\sqrt{8 - 2 \cdot 2} = \sqrt 4 = 2. Domain: 8−2x≥08 - 2x \ge 0, so x≤4x \le 4, that is (−∞,4](-\infty, 4]: the half-parabola now opens to the LEFT. Reading 8−2x8 - 2x as a shift of 88 is the trap: the shift is 44, visible only after factoring.

d) 1+2x−1=(x−1)+2x−1=x+1x−11 + \frac{2}{x - 1} = \frac{(x - 1) + 2}{x - 1} = \frac{x + 1}{x - 1}. From y=1xy = \frac{1}{x}: the outside factor 22 stretches vertically by 22, the inside x−1x - 1 shifts RIGHT 11, and the outside +1+1 shifts up 11. The hyperbola, centred at (0,0)(0, 0) with its two branches in the first and third quadrants, is now centred at (1,1)(1, 1). Domain: x≠1x \ne 1, where the denominator vanishes. Range: 2x−1\frac{2}{x - 1} takes every nonzero value but never 00, so y=1+2x−1y = 1 + \frac{2}{x - 1} takes every value except 11. Check: x=3x = 3 gives 42=2\frac{4}{2} = 2 and 1+22=21 + \frac{2}{2} = 2. Solving x+1x−1=1\frac{x + 1}{x - 1} = 1 gives x+1=x−1x + 1 = x - 1, impossible, which confirms that 11 is not in the range.

e) The corner of y=∣x∣y = |x| is at (0,0)(0, 0); here it is at (−1,3)(-1, 3), so h=−1h = -1 and k=3k = 3: g(x)=a∣x+1∣+3g(x) = a|x + 1| + 3. The graph opens downward, so a<0a < 0; the point (1,−1)(1, -1) gives −1=a⋅2+3-1 = a \cdot 2 + 3, so a=−2a = -2. Check with the third point: g(−3)=−2⋅2+3=−1g(-3) = -2 \cdot 2 + 3 = -1, as on the figure. So g(x)=−2∣x+1∣+3g(x) = -2|x + 1| + 3: the graph of ∣x∣|x| stretched vertically by 22, reflected, shifted left 11 and up 33. Intercepts: −2∣x+1∣+3=0-2|x + 1| + 3 = 0 gives ∣x+1∣=32|x + 1| = \frac{3}{2}, so x=12x = \frac{1}{2} or x=−52x = -\frac{5}{2}. Writing ∣x−1∣|x - 1| for a corner at x=−1x = -1 is the sign error of the chapter; checking the corner in the formula takes two seconds.

Part B: problems and reasoning (/50)

Exercise 6: Composition: the inner function first, then its outputs

The composition (f∘g)(x)=f(g(x))(f \circ g)(x) = f(g(x)) applies gg FIRST, then ff to the result. Its domain is not read on the simplified formula: xx must be in the domain of gg, AND g(x)g(x) must be in the domain of ff. Composition is not multiplication, and in general f∘g≠g∘ff \circ g \ne g \circ f.

Parts a) and b) use u(x)=xu(x) = \sqrt x and v(x)=1x−2v(x) = \frac{1}{x - 2}. Part d) uses the two functions ff and gg whose graphs, made of line segments, are shown in the figure.

1234567812345y = f(x)y = g(x)x
  • a) Find u∘vu \circ v and v∘uv \circ u, each with its domain.
  • b) Find v∘vv \circ v and its domain. A student gives the domain x≠52x \ne \frac{5}{2} from the simplified formula: what is missing?
  • c) Write H(x)=(3−x)4H(x) = (3 - \sqrt x)^4 as a composition of two functions, then of three. Is the decomposition unique?
  • d) From the figure, find f(g(3))f(g(3)), g(f(3))g(f(3)), (f∘f)(1)(f \circ f)(1) and (g∘g)(6)(g \circ g)(6), then solve f(g(x))=0f(g(x)) = 0.
  • e) With p(x)=2x+3p(x) = 2x + 3, find qq such that (p∘q)(x)=2x2+5(p \circ q)(x) = 2x^2 + 5. Then find rr such that r(x+4)=x2+8x+20r(x + 4) = x^2 + 8x + 20 for every xx.

Type your answers, the page tells you right or wrong 0/13

a)
Domain of u∘vu \circ v ,
b)
c)
d)
e)
Show the solution

Answers

  • a) (u∘v)(x)=1x−2(u \circ v)(x) = \frac{1}{\sqrt{x - 2}}, domain (2,∞)(2, \infty); (v∘u)(x)=1x−2(v \circ u)(x) = \frac{1}{\sqrt x - 2}, domain [0,4)∪(4,∞)[0, 4) \cup (4, \infty)
  • b) (v∘v)(x)=x−25−2x(v \circ v)(x) = \frac{x - 2}{5 - 2x}, domain R∖{2,52}\mathbb{R} \setminus \{2, \frac{5}{2}\}: x=2x = 2 is excluded by the inner vv
  • c) H=F∘GH = F \circ G with G(x)=3−xG(x) = 3 - \sqrt x, F(x)=x4F(x) = x^4; H=F∘M∘uH = F \circ M \circ u with M(x)=3−xM(x) = 3 - x; not unique
  • d) f(g(3))=2f(g(3)) = 2, g(f(3))=2g(f(3)) = 2, (f∘f)(1)=0(f \circ f)(1) = 0, (g∘g)(6)=2(g \circ g)(6) = 2; f(g(x))=0f(g(x)) = 0 for x=1x = 1 and x=5x = 5
  • e) q(x)=x2+1q(x) = x^2 + 1; r(t)=t2+4r(t) = t^2 + 4

a) (u∘v)(x)=u(1x−2)=1x−2(u \circ v)(x) = u\left(\frac{1}{x - 2}\right) = \sqrt{\frac{1}{x - 2}}. Inner function first: x≠2x \ne 2. Then its output must be a legal input of  \sqrt{\ }: 1x−2≥0\frac{1}{x - 2} \ge 0, and a quotient with numerator 11 is positive exactly when x−2>0x - 2 > 0. Domain (2,∞)(2, \infty), and u∘v=1x−2u \circ v = \frac{1}{\sqrt{x - 2}} there. (v∘u)(x)=v(x)=1x−2(v \circ u)(x) = v(\sqrt x) = \frac{1}{\sqrt x - 2}. Inner: x≥0x \ge 0. Outer: its input x\sqrt x must not be 22, so x≠4x \ne 4. Domain [0,4)∪(4,∞)[0, 4) \cup (4, \infty). The two compositions are different functions with different domains: at x=9x = 9, u(v(9))=17u(v(9)) = \frac{1}{\sqrt 7} while v(u(9))=13−2=1v(u(9)) = \frac{1}{3 - 2} = 1.

b) (v∘v)(x)=11x−2−2(v \circ v)(x) = \frac{1}{\frac{1}{x - 2} - 2}. Multiply numerator and denominator by x−2x - 2: x−21−2(x−2)=x−25−2x\frac{x - 2}{1 - 2(x - 2)} = \frac{x - 2}{5 - 2x}. Domain, inside out: the inner vv needs x≠2x \ne 2; the outer vv needs its input 1x−2\frac{1}{x - 2} different from 22, that is x−2≠12x - 2 \ne \frac{1}{2}, x≠52x \ne \frac{5}{2}. So the domain is R∖{2,52}\mathbb{R} \setminus \left\{2, \frac{5}{2}\right\}. The simplified formula x−25−2x\frac{x - 2}{5 - 2x} happily gives 00 at x=2x = 2, but v(2)v(2) does not exist, so v(v(2))v(v(2)) does not exist either: multiplying by x−2x - 2 was legal only because x≠2x \ne 2 had already been assumed. The domain comes from the construction, never from the final formula.

c) Read HH from the inside out, as a calculator would compute it: take x\sqrt x, subtract it from 33, raise to the fourth power. Two layers: G(x)=3−xG(x) = 3 - \sqrt x and F(x)=x4F(x) = x^4 give F(G(x))=(3−x)4F(G(x)) = (3 - \sqrt x)^4. Three layers: u(x)=xu(x) = \sqrt x, M(x)=3−xM(x) = 3 - x, F(x)=x4F(x) = x^4, and H=F∘M∘uH = F \circ M \circ u, with uu applied FIRST although it is written LAST. The decomposition is not unique: F2(x)=x2F_2(x) = x^2 and G2(x)=(3−x)2G_2(x) = (3 - \sqrt x)^2 also work, since ((3−x)2)2=(3−x)4\left((3 - \sqrt x)^2\right)^2 = (3 - \sqrt x)^4. Writing H=u∘M∘FH = u \circ M \circ F in the reading order of the formula computes 3−x4\sqrt{3 - x^4}, another function.

d) Read the inner value on the inner graph, then use it as an input on the outer graph. g(3)=4g(3) = 4, the top of the orange graph, and f(4)=2f(4) = 2, so f(g(3))=2f(g(3)) = 2. f(3)=1f(3) = 1 and g(1)=2g(1) = 2, so g(f(3))=2g(f(3)) = 2. f(1)=2f(1) = 2 and f(2)=0f(2) = 0, so (f∘f)(1)=0(f \circ f)(1) = 0. g(6)=1g(6) = 1 and g(1)=2g(1) = 2, so (g∘g)(6)=2(g \circ g)(6) = 2. To solve f(g(x))=0f(g(x)) = 0, go from the outside in: f(t)=0f(t) = 0 only for t=2t = 2 (the lowest point of the blue graph), so we need g(x)=2g(x) = 2. On the orange graph, g(x)=x+1g(x) = x + 1 on [0,3][0, 3] and g(x)=7−xg(x) = 7 - x on [3,6][3, 6], so g(x)=2g(x) = 2 for x=1x = 1 and x=5x = 5. Both solutions are needed: stopping at the first intersection costs half the question.

e) p(q(x))=2q(x)+3p(q(x)) = 2q(x) + 3 must equal 2x2+52x^2 + 5, so 2q(x)=2x2+22q(x) = 2x^2 + 2 and q(x)=x2+1q(x) = x^2 + 1. Check: p(x2+1)=2x2+2+3p(x^2 + 1) = 2x^2 + 2 + 3. For rr, name the inner output: t=x+4t = x + 4, so x=t−4x = t - 4, and r(t)=(t−4)2+8(t−4)+20=t2−8t+16+8t−32+20=t2+4r(t) = (t - 4)^2 + 8(t - 4) + 20 = t^2 - 8t + 16 + 8t - 32 + 20 = t^2 + 4. Faster, complete the square: x2+8x+20=(x+4)2+4x^2 + 8x + 20 = (x + 4)^2 + 4, which shows at once that rr squares its input and adds 44. Check at x=0x = 0: r(4)=20r(4) = 20, and 0+0+20=200 + 0 + 20 = 20. The error to avoid is r(x)=x2+8x+20r(x) = x^2 + 8x + 20 itself: that is r∘(x+4)r \circ (x + 4), not rr.

Exercise 7: Linear, power and rational models: choosing the model and its domain

A mathematical model is a function chosen to describe data. Three families of Stewart's catalogue are used here. LINEAR, y=mx+by = mx + b: equal steps in xx add a CONSTANT amount to yy, the slope, which has units. POWER, y=kxpy = kx^p: multiplying xx by 22 always MULTIPLIES yy by 2p2^p. RATIONAL, such as y=kxy = \frac{k}{x}, the power p=−1p = -1. A model also has a domain, where it describes reality, and it is wrong outside it.

The figure shows three data points (a,T)(a, T), where aa is the average distance of a body from its star (in astronomical units, AU) and TT its period of revolution (in years), with the power curve through them.

1234567891051015202530(1, 1)(4, 8)(9, 27)T = k·aᵖa (AU)T (years)
  • a) The air temperature is 15 ∘C15\,^\circ\mathrm{C} at ground level and 2 ∘C2\,^\circ\mathrm{C} at an altitude of 22 km. Assuming a linear model T(h)T(h), find it, give the meaning and the units of its slope, and find the altitude where the temperature is −50 ∘C-50\,^\circ\mathrm{C}.
  • b) Find kk and pp in T=kapT = ka^p from the figure, then the period of a body at 1616 AU and the distance of a body whose period is 125125 years.
  • c) Decide which of these two tables comes from a linear model and which from a power model, and find each formula. Table 1: x=1,2,3,4x = 1, 2, 3, 4 gives y=5,8,11,14y = 5, 8, 11, 14. Table 2: x=1,2,4,8x = 1, 2, 4, 8 gives y=2,16,128,1024y = 2, 16, 128, 1024.
  • d) At constant temperature, the pressure of a gas is P=240VP = \frac{240}{V} kPa, for a volume VV in litres (Boyle's law). Compute PP for V=8V = 8, 44 and 22. What happens to PP when VV is tripled? Give the domain of the model and name its family.
  • e) Use the model of a) at h=50h = 50 km, and explain what the result says about the domain of the model.

Type your answers, the page tells you right or wrong 0/17

a)
b)
c)
d)
e)
Show the solution

Answers

  • a) T(h)=15−6.5hT(h) = 15 - 6.5h; slope −6.5 ∘C-6.5\,^\circ\mathrm{C} per km; h=10h = 10 km
  • b) k=1k = 1, p=32p = \frac{3}{2}; T(16)=64T(16) = 64 years; a=25a = 25 AU
  • c) Table 1 linear, y=3x+2y = 3x + 2; Table 2 power, y=2x3y = 2x^3
  • d) 3030, 6060, 120120 kPa; PP is divided by 33; domain V>0V > 0; power function with p=−1p = -1 (rational)
  • e) T(50)=−310 ∘CT(50) = -310\,^\circ\mathrm{C}, below absolute zero: the model is only valid on a limited range of altitudes.

a) Two points (0,15)(0, 15) and (2,2)(2, 2): slope m=2−152−0=−6.5m = \frac{2 - 15}{2 - 0} = -6.5, intercept 1515, so T(h)=15−6.5hT(h) = 15 - 6.5h. The slope is a RATE with units, ∘C^\circ\mathrm{C} per km: the temperature drops by 6.56.5 degrees for each kilometre climbed. The intercept 15 ∘C15\,^\circ\mathrm{C} is the ground temperature. For T=−50T = -50: 15−6.5h=−5015 - 6.5h = -50, so 6.5h=656.5h = 65 and h=10h = 10 km. An answer h=−10h = -10 comes from a sign error on the slope, and a negative altitude should trigger a second look.

b) The point (1,1)(1, 1) gives 1=k⋅1p=k1 = k \cdot 1^p = k, so k=1k = 1. The point (4,8)(4, 8) then gives 4p=84^p = 8; write both sides as powers of 22: 22p=232^{2p} = 2^3, so 2p=32p = 3 and p=32p = \frac{3}{2}. Check with the third point: 93/2=(9)3=279^{3/2} = (\sqrt 9)^3 = 27. This is Kepler's third law, T=a3/2T = a^{3/2}. At 1616 AU: T=163/2=43=64T = 16^{3/2} = 4^3 = 64 years. For T=125T = 125: a3/2=125a^{3/2} = 125, so a=1253=5\sqrt a = \sqrt[3]{125} = 5 and a=25a = 25 AU. Take the root first and square after: numbers stay small, and no calculator is needed. The curve is not a straight line: a linear fit through (1,1)(1, 1) and (4,8)(4, 8) would predict T(9)=1+73⋅8=593T(9) = 1 + \frac{7}{3} \cdot 8 = \frac{59}{3}, far from 2727.

c) Table 1: equal steps of xx (+1+1) give equal steps of yy (+3+3): constant differences, linear, slope 33, and y(1)=5y(1) = 5 gives y=3x+2y = 3x + 2. Table 2: the xx-values DOUBLE, and each time yy is multiplied by the same factor, 162=12816=1024128=8\frac{16}{2} = \frac{128}{16} = \frac{1024}{128} = 8. For a power function kxpkx^p, doubling xx multiplies yy by 2p2^p, so 2p=82^p = 8, p=3p = 3, and y(1)=2y(1) = 2 gives k=2k = 2: y=2x3y = 2x^3. Check: 2⋅83=10242 \cdot 8^3 = 1024. The differences of Table 2 (1414, 112112, 896896) are not constant, and the ratios of Table 1 (85\frac{8}{5}, 118\frac{11}{8}) are not constant: each test identifies ONE family and rejects the other.

d) P(8)=30P(8) = 30, P(4)=60P(4) = 60, P(2)=120P(2) = 120 kPa: halving the volume doubles the pressure. If VV is tripled, P=2403V=13⋅240VP = \frac{240}{3V} = \frac{1}{3} \cdot \frac{240}{V}: the pressure is divided by 33, which is the power rule 3−13^{-1}. The formula accepts every V≠0V \ne 0, but a volume is positive, so the domain of the MODEL is V>0V > 0, only the branch of the hyperbola in the first quadrant. The family: P=240V−1P = 240V^{-1} is a power function with p=−1p = -1, which is also the simplest rational function. Algebraic domain and physical domain differ, and a model answer must give the second.

e) T(50)=15−6.5×50=15−325=−310 ∘CT(50) = 15 - 6.5 \times 50 = 15 - 325 = -310\,^\circ\mathrm{C}. No temperature is below absolute zero, about −273 ∘C-273\,^\circ\mathrm{C}, so the prediction is impossible. The formula is defined for every real hh, but the MODEL is not: the linear decrease describes the lowest layer of the atmosphere only, roughly the first 1111 km, and above it the temperature stops decreasing. Using a model outside the range of its data is called extrapolation, and the linear model extrapolated here does not bend where reality does. The domain of a model is part of the model.

Exercise 8: Five statements to correct

Each statement below was written by a student on a MATH 140 assignment, and each is false. Say what is wrong, give a counterexample with numbers, and write the correct statement.

  • a) The graph of y=f(x+3)y = f(x + 3) is the graph of ff shifted 33 units to the right.
  • b) For any two functions, f∘g=g∘ff \circ g = g \circ f.
  • c) If ff and gg are both odd, then f∘gf \circ g is even.
  • d) To obtain the graph of y=f(2x)y = f(2x), stretch the graph of ff horizontally by a factor of 22.
  • e) The graphs of y=−f(x)y = -f(x) and y=f(−x)y = f(-x) are the same, since both put a minus sign on ff.

Type your answers, the page tells you right or wrong 0/5

a)
b)
c)
d)
e)
Show the solution

Answers

  • a) False: (x+3)2(x + 3)^2 has its vertex at x=−3x = -3. f(x+3)f(x + 3) shifts LEFT 33.
  • b) False: f(x)=x2f(x) = x^2, g(x)=x+1g(x) = x + 1 give 4≠24 \ne 2 at x=1x = 1.
  • c) False: f(x)=g(x)=xf(x) = g(x) = x give f∘g=xf \circ g = x, odd. Odd ∘\circ odd is odd.
  • d) False: the zeros ±2\pm 2 of x2−4x^2 - 4 become ±1\pm 1. f(2x)f(2x) COMPRESSES by 22.
  • e) False: −x2≠x2-x^2 \ne x^2. They coincide exactly when ff is odd.

a) FALSE. Take f(x)=x2f(x) = x^2, vertex at 00: f(x+3)=(x+3)2f(x + 3) = (x + 3)^2 has its vertex where the INSIDE is 00, at x=−3x = -3, three units to the LEFT. The inside runs backwards: the new graph must reach, at xx, the value that ff had at x+3x + 3, three units further right, so it is behind by 33. Correct statement: y=f(x+3)y = f(x + 3) is the graph of ff shifted 33 units to the left, and y=f(x−3)y = f(x - 3) is shifted 33 units to the right. The check that settles it in five seconds: where does the inside vanish?

b) FALSE. With f(x)=x2f(x) = x^2 and g(x)=x+1g(x) = x + 1: (f∘g)(x)=(x+1)2(f \circ g)(x) = (x + 1)^2 and (g∘f)(x)=x2+1(g \circ f)(x) = x^2 + 1. At x=1x = 1: 44 and 22. Squaring then adding is not adding then squaring. Correct statement: in general f∘g≠g∘ff \circ g \ne g \circ f; they may agree for special pairs, such as f(x)=2xf(x) = 2x and g(x)=3xg(x) = 3x, which both give 6x6x, but the order must be read from the inside: in f(g(x))f(g(x)), gg acts first.

c) FALSE. With f(x)=g(x)=xf(x) = g(x) = x, both odd, (f∘g)(x)=x(f \circ g)(x) = x, which is odd, not even. In general, f(g(−x))=f(−g(x))=−f(g(x))f(g(-x)) = f(-g(x)) = -f(g(x)): the minus sign produced by gg passes through ff, which sends it outside. So odd ∘\circ odd is ODD. What the student remembered belongs to the PRODUCT: (fg)(−x)=(−f(x))(−g(x))=f(x)g(x)(fg)(-x) = (-f(x))(-g(x)) = f(x)g(x), so the product of two odd functions is even, as x⋅x3=x4x \cdot x^3 = x^4 shows. Composition and product obey different sign rules, because composition is not multiplication.

d) FALSE. Take f(x)=x2−4f(x) = x^2 - 4, with zeros ±2\pm 2. Then f(2x)=4x2−4f(2x) = 4x^2 - 4, with zeros ±1\pm 1: the graph got NARROWER. A point (a,b)(a, b) of ff goes where 2x=a2x = a, at (a2,b)\left(\frac{a}{2}, b\right): every xx-coordinate is halved. Correct statement: y=f(2x)y = f(2x) compresses the graph horizontally by a factor 22; it is y=f(x2)y = f\left(\frac{x}{2}\right) that stretches it by 22. Outside, 2f(x)2f(x) does stretch vertically by 22: the same number has opposite effects inside and outside.

e) FALSE. For f(x)=x2f(x) = x^2: −f(x)=−x2-f(x) = -x^2, a parabola opening downward, and f(−x)=(−x)2=x2f(-x) = (-x)^2 = x^2, the same parabola as ff. At x=1x = 1: −1-1 and 11. The minus sign OUTSIDE acts on yy (reflection in the xx-axis); INSIDE it acts on xx (reflection in the yy-axis). Correct statement: y=−f(x)y = -f(x) and y=f(−x)y = f(-x) are the reflections of the graph of ff in the xx-axis and in the yy-axis; they coincide exactly when f(−x)=−f(x)f(-x) = -f(x) for all xx, that is when ff is odd, as x3x^3 shows.

Exercise 9: Income tax by brackets: a piecewise linear function

A country taxes income by brackets: 0%0\% on the first 10 00010\,000 dollars, 20%20\% on the part of the income between 10 00010\,000 and 40 00040\,000 dollars, and 30%30\% on the part above 40 00040\,000 dollars. Each rate applies only to the PART of the income that lies in its bracket. Let T(I)T(I) be the tax, in dollars, on an income of II dollars.

The figure shows the graph of TT, with both axes in thousands of dollars. No calculator: every value is exact.

102030405060702468101214160%20%30%(40, 6)income I (thousands of dollars)tax T
  • a) Compute T(8000)T(8000), T(25 000)T(25\,000) and T(60 000)T(60\,000).
  • b) Write T(I)T(I) as a piecewise formula for I≥0I \ge 0, in the form mI+bmI + b on each piece, and check that the pieces agree at both thresholds.
  • c) A worker earning 39 00039\,000 dollars is offered a raise to 41 00041\,000 dollars and fears that the higher bracket will leave him with less. Write his income after tax N(I)=I−T(I)N(I) = I - T(I) as a piecewise formula, and settle the question.
  • d) The average tax rate is A(I)=T(I)IA(I) = \frac{T(I)}{I}. Compute A(60 000)A(60\,000) and compare it with the rate of the last bracket. At what income is the average rate 15%15\%?
  • e) A person paid 90009000 dollars of tax. What was the income? Treat every piece.

Type your answers, the page tells you right or wrong 0/14

a)
b)
c)
d)
e)
Show the solution

Answers

  • a) 00, 30003000 and 12 00012\,000 dollars
  • b) T(I)=0T(I) = 0 on [0,10 000][0, 10\,000], 0.2I−20000.2I - 2000 on (10 000,40 000](10\,000, 40\,000], 0.3I−60000.3I - 6000 for I>40 000I > 40\,000; values 00 and 60006000 at the thresholds
  • c) N(I)=IN(I) = I, 0.8I+20000.8I + 2000, 0.7I+60000.7I + 6000; N(39 000)=33 200<N(41 000)=34 700N(39\,000) = 33\,200 < N(41\,000) = 34\,700
  • d) A(60 000)=20%A(60\,000) = 20\%, below 30%30\%; A(I)=15%A(I) = 15\% at I=40 000I = 40\,000
  • e) I=50 000I = 50\,000 dollars (55 00055\,000 rejected in the 20%20\% piece)

a) 80008000 is in the first bracket: T(8000)=0T(8000) = 0. For 25 00025\,000, the first 10 00010\,000 are free and the remaining 15 00015\,000 are taxed at 20%20\%: T(25 000)=0.2×15 000=3000T(25\,000) = 0.2 \times 15\,000 = 3000. For 60 00060\,000: 00 on the first 10 00010\,000, 20%20\% of the 30 00030\,000 between 10 00010\,000 and 40 00040\,000, that is 60006000, and 30%30\% of the 20 00020\,000 above 40 00040\,000, that is 60006000: T(60 000)=12 000T(60\,000) = 12\,000. The classic error is 30%30\% of the whole income, 18 00018\,000 dollars: the top rate applies only to the top part. On the figure, T(60)=12T(60) = 12 in thousands.

b) First piece, 0≤I≤10 0000 \le I \le 10\,000: T(I)=0T(I) = 0. Second, 10 000<I≤40 00010\,000 < I \le 40\,000: T(I)=0.2(I−10 000)=0.2I−2000T(I) = 0.2(I - 10\,000) = 0.2I - 2000. Third, I>40 000I > 40\,000: the full second bracket, 0.2×30 000=60000.2 \times 30\,000 = 6000, plus 30%30\% of the excess, T(I)=6000+0.3(I−40 000)=0.3I−6000T(I) = 6000 + 0.3(I - 40\,000) = 0.3I - 6000. At 10 00010\,000: the first formula gives 00, the second 2000−2000=02000 - 2000 = 0. At 40 00040\,000: 8000−2000=60008000 - 2000 = 6000 and 12 000−6000=600012\,000 - 6000 = 6000. The formulas AGREE at each threshold, so the graph has no jump: it only bends, and the slope of each piece is its rate, 00, 0.20.2, 0.30.3. Writing 0.3I0.3I on the third piece, without the −6000-6000, would create a jump of 60006000 dollars at 40 00040\,000.

c) N(I)=I−T(I)N(I) = I - T(I) piece by piece: N(I)=IN(I) = I on [0,10 000][0, 10\,000], N(I)=I−(0.2I−2000)=0.8I+2000N(I) = I - (0.2I - 2000) = 0.8I + 2000 on (10 000,40 000](10\,000, 40\,000], and N(I)=I−(0.3I−6000)=0.7I+6000N(I) = I - (0.3I - 6000) = 0.7I + 6000 for I>40 000I > 40\,000. Then N(39 000)=31 200+2000=33 200N(39\,000) = 31\,200 + 2000 = 33\,200 and N(41 000)=28 700+6000=34 700N(41\,000) = 28\,700 + 6000 = 34\,700: the raise gains him 15001500 dollars after tax. More generally, every piece of NN has a POSITIVE slope (11, 0.80.8, 0.70.7) and the pieces agree at the thresholds, so NN always increases: under bracket taxation, earning more never leaves you with less. Only the extra dollars above 40 00040\,000 are taxed at 30%30\%, and each of them still leaves 7070 cents.

d) A(60 000)=12 00060 000=15=20%A(60\,000) = \frac{12\,000}{60\,000} = \frac{1}{5} = 20\%, well below the 30%30\% rate of the last bracket: the lower brackets pull the average down. The rate of the bracket is the MARGINAL rate, the slope of TT; the average rate is the slope of the line from the origin to the point of the graph. For A(I)=15%A(I) = 15\%, try the second piece: 0.2I−2000I=0.15\frac{0.2I - 2000}{I} = 0.15 gives 0.05I=20000.05I = 2000, I=40 000I = 40\,000, which is in (10 000,40 000](10\,000, 40\,000]: kept. Check: T(40 000)=6000T(40\,000) = 6000 and 600040 000=0.15\frac{6000}{40\,000} = 0.15. On the first piece A=0A = 0; on the third, 0.3I−6000=0.15I0.3I - 6000 = 0.15I gives I=40 000I = 40\,000 again, not in I>40 000I > 40\,000. The answer is 40 00040\,000 dollars, once.

e) Solve T(I)=9000T(I) = 9000 on each piece, then check the candidate against its piece. First piece: 0=90000 = 9000, impossible. Second: 0.2I−2000=90000.2I - 2000 = 9000 gives I=55 000I = 55\,000, NOT in (10 000,40 000](10\,000, 40\,000]: rejected (the formula of the second bracket does not apply to that income). Third: 0.3I−6000=90000.3I - 6000 = 9000 gives 0.3I=15 0000.3I = 15\,000, I=50 000>40 000I = 50\,000 > 40\,000: kept. Check with the brackets: 6000+0.3×10 000=90006000 + 0.3 \times 10\,000 = 9000. The income was 50 00050\,000 dollars. A shortcut exists: TT reaches only 60006000 at the end of the second bracket, and 9000>60009000 > 6000, so only the third piece can answer.

Exercise 10: A final exam problem: coupon, discount and free delivery, composed

An online store runs two promotions on a cart worth pp dollars: a sale of 20%20\% off, D(p)=0.8pD(p) = 0.8p, and a coupon of 1010 dollars off, C(q)=q−10C(q) = q - 10, valid when the amount it applies to is at least 1010 dollars. Delivery costs 88 dollars when the amount due after reductions is below 5050 dollars, and is free from 5050 dollars on.

The figure shows the amount paid A(q)A(q) as a function of the amount qq due after reductions: A(q)=q+8A(q) = q + 8 for 0≤q<500 \le q < 50 and A(q)=qA(q) = q for q≥50q \ge 50.

1020304050607080102030405060708090shipping: 8 dollarsfree shippingq (dollars, after reductions)A (dollars)
  • a) Compute D(C(100))D(C(100)) and C(D(100))C(D(100)). Find formulas for D∘CD \circ C and C∘DC \circ D, and show that one order is always cheaper for the customer.
  • b) The store applies the sale first, then the coupon, then delivery: the total is T(p)=A(C(D(p)))T(p) = A(C(D(p))). Find the domain of C∘DC \circ D, then write T(p)T(p) as a piecewise formula.
  • c) Compare T(70)T(70) and T(75)T(75). For which carts p<75p < 75 does the customer pay more than with a cart of 7575 dollars?
  • d) A sales tax of 5%5\% is added on the total: F(p)=1.05 T(p)F(p) = 1.05\,T(p). Compute F(70)F(70) and F(75)F(75). Which transformation takes the graph of TT to that of FF, and what happens to the drop at p=75p = 75?
  • e) A second store uses the same reductions but decides delivery on the cart value pp BEFORE reductions: free from p=50p = 50 on. Write its total T2(p)T_2(p) and compare T2(60)T_2(60) with T(60)T(60).

Type your answers, the page tells you right or wrong 0/15

a)
b)
Domain of C∘DC \circ D ,
c)
Carts p<75p < 75 that pay more ,
d)
e)
Show the solution

Answers

  • a) 7272 and 7070 dollars; (D∘C)(p)=0.8p−8(D \circ C)(p) = 0.8p - 8, (C∘D)(p)=0.8p−10(C \circ D)(p) = 0.8p - 10: sale first is always 22 dollars cheaper
  • b) Domain p≥12.5p \ge 12.5; T(p)=0.8p−2T(p) = 0.8p - 2 on [12.5,75)[12.5, 75), 0.8p−100.8p - 10 for p≥75p \ge 75
  • c) T(70)=54>T(75)=50T(70) = 54 > T(75) = 50; carts with 65<p<7565 < p < 75
  • d) F(70)=56.70F(70) = 56.70, F(75)=52.50F(75) = 52.50 dollars; vertical stretch by 1.051.05; the drop grows from 88 to 8.408.40 dollars and stays at p=75p = 75
  • e) T2(p)=0.8p−2T_2(p) = 0.8p - 2 on [12.5,50)[12.5, 50), 0.8p−100.8p - 10 for p≥50p \ge 50; T2(60)=38<T(60)=46T_2(60) = 38 < T(60) = 46

a) D(C(100))=D(90)=72D(C(100)) = D(90) = 72: coupon first, then sale. C(D(100))=C(80)=70C(D(100)) = C(80) = 70: sale first, then coupon. In general (D∘C)(p)=0.8(p−10)=0.8p−8(D \circ C)(p) = 0.8(p - 10) = 0.8p - 8 and (C∘D)(p)=0.8p−10(C \circ D)(p) = 0.8p - 10, so (D∘C)(p)−(C∘D)(p)=2(D \circ C)(p) - (C \circ D)(p) = 2 for every cart: applying the sale FIRST always saves 22 dollars. The reason is visible in the formulas: when the coupon acts first, the sale also shrinks the coupon, to 0.8×10=80.8 \times 10 = 8 dollars. Composition order is not a detail, it is money.

b) C∘DC \circ D: the inner function DD accepts every p≥0p \ge 0; the coupon then needs D(p)=0.8p≥10D(p) = 0.8p \ge 10, so p≥12.5p \ge 12.5. The domain is [12.5,∞)[12.5, \infty), and q=C(D(p))=0.8p−10≥0q = C(D(p)) = 0.8p - 10 \ge 0 there. Then AA asks which piece qq belongs to: q<50q < 50 exactly when 0.8p−10<500.8p - 10 < 50, that is p<75p < 75. So T(p)=(0.8p−10)+8=0.8p−2T(p) = (0.8p - 10) + 8 = 0.8p - 2 for 12.5≤p<7512.5 \le p < 75, and T(p)=0.8p−10T(p) = 0.8p - 10 for p≥75p \ge 75. The threshold of delivery is 5050 in the variable qq, but 7575 in the variable pp: the outer function's breakpoint is pulled back through the inner function, the inside-out reading of the whole chapter. The solution figure shows TT, with its drop at p=75p = 75.

c) T(70)=56−2=54T(70) = 56 - 2 = 54 and T(75)=60−10=50T(75) = 60 - 10 = 50: the larger cart costs 44 dollars LESS. For p<75p < 75 the total 0.8p−20.8p - 2 exceeds 5050 when 0.8p>520.8p > 52, that is p>65p > 65. So every cart strictly between 6565 and 7575 dollars pays more than a cart of 7575: those customers should add items up to 7575 dollars. At p=65p = 65 exactly, T(65)=50T(65) = 50, a tie. On the figure of the solution, this is the part of the first piece above the dashed level T=50T = 50. A piecewise price with a jump always creates such a zone, and finding it is a question of solving an inequality on ONE piece and checking it against the interval of that piece.

d) F(70)=1.05×54=56.70F(70) = 1.05 \times 54 = 56.70 and F(75)=1.05×50=52.50F(75) = 1.05 \times 50 = 52.50 dollars. The factor 1.051.05 is OUTSIDE TT: it acts on yy only, a vertical stretch by 1.051.05, and leaves every xx-coordinate in place. So the drop is still at p=75p = 75, where the inside decides; its size, 58−50=858 - 50 = 8 dollars on TT (the first formula evaluated at 7575 gives 0.8×75−2=580.8 \times 75 - 2 = 58, the level just before the drop), becomes 1.05×8=8.401.05 \times 8 = 8.40 dollars on FF. The zone of part c) does not move either: F(p)>F(75)F(p) > F(75) exactly when T(p)>T(75)T(p) > T(75), since multiplying by a positive number keeps inequalities.

e) Now the delivery rule reads the variable pp itself: T2(p)=(0.8p−10)+8=0.8p−2T_2(p) = (0.8p - 10) + 8 = 0.8p - 2 for 12.5≤p<5012.5 \le p < 50, and T2(p)=0.8p−10T_2(p) = 0.8p - 10 for p≥50p \ge 50. The drop moves from p=75p = 75 to p=50p = 50. For a cart of 6060 dollars, q=38q = 38: the first store charges delivery, T(60)=46T(60) = 46, the second does not, T2(60)=38T_2(60) = 38. Same three rules, different totals, because the threshold is applied to a different INPUT. When a price has several rules, the first question is always which quantity each rule reads, and that is exactly what composition writes down.

1020304050607080901001020304050607080p = 65T = 50(75, 50)cart p (dollars)total T

© Ahmed Squalli Houssaini. Exercise set published at www.letuteurscientifique.ca/en/exercices/math140-functions-review. Free for personal and classroom use; republishing it elsewhere requires written permission (legal notice).

See also

Struggling with MATH 140?

I tutor first-year calculus at McGill and Concordia, in English or in French, in Montreal or online. Get in touch for a first session.

Site by Studio Squalli