Skip to main content
\( \definecolor{colordef}{RGB}{249,49,84} \definecolor{colorprop}{RGB}{18,102,241} \)
International Baccalaureate · DP 1: Analysis and Approaches HL

Transformations of Functions

⌚ ~130 min ▢ 14 blocks ✓ 63 exercises ➣ Prerequisites : Functions, Functions
I Translations, Stretches and Reflections
I.1 Translations
Recappreviously studied — click to reread
Function transformations take a basic “parent” function and modify it to create a new, related function. A sequence of transformations can translate, stretch or reflect the graph of the parent function.
Understanding these transformations makes it possible to predict the graph of a more complicated function from a simpler one, and to model real-world phenomena by adjusting a basic function to fit observed data. The transformations are vertical and horizontal translations, stretches (a stretch with a scale factor between \(0\) and \(1\) is also called a compression) and reflections.
Report an error
Definition — Vertical Translation by Adding a Constant
A vertical translation shifts the graph of a function up or down. The transformation is defined by:$$g(x) = f(x) + k$$ This transformation maps a point \((x, y)\) on the graph of \(f\) to a new point \((x, y+k)\) on the graph of \(g\).
  • If \(k>0\), the graph is translated \(\boldsymbol{k}\) units up.
  • If \(k<0\), the graph is translated \(\boldsymbol{|k|}\) units down.
Figure
Definition — Horizontal Translation
A horizontal translation shifts the graph of a function left or right. The transformation is defined by:$$g(x) = f(x-h)$$ This transformation maps a point \((x, y)\) on the graph of \(f\) to a new point \((x+h, y)\) on the graph of \(g\).
  • If \(h>0\), the graph is translated \(\boldsymbol{h}\) units to the right.
  • If \(h<0\), the graph is translated \(\boldsymbol{|h|}\) units to the left.
Figure
Example
From the graph of \(f(x) = x^2\), plot \(g(x) = f(x) - 3\) and \(h(x) = f(x + 1)\).
  • \(g(x) = f(x) - 3\): every output is lowered by \(3\) — a vertical translation \(3\) units down. The vertex \((0, 0)\) moves to \((0, -3)\).
  • \(h(x) = f(x + 1)\): the input \(x = -1\) now plays the role \(0\) used to play — a horizontal translation \(1\) unit to the left. The vertex \((0, 0)\) moves to \((-1, 0)\).
Figure
Translation by a Vector
The vertical and the horizontal translations combine into one. The translation by the vector \(\begin{pmatrix} h \\ k \end{pmatrix}\) moves each point \((x, y)\) of the graph of \(f\) to the point \((X, Y) = (x + h, y + k)\). Since \(x = X - h\) and \(Y = f(x) + k\), the image points satisfy \(Y = f(X - h) + k\); writing \((x, y)\) again for the coordinates, the image of the graph of \(y = f(x)\) is the graph of $$y = f(x - h) + k.$$ For example, the translation by the vector \(\begin{pmatrix} 3 \\ -2 \end{pmatrix}\) maps the graph of \(y = x^2\) onto the graph of \(y = (x - 3)^2 - 2\), and its vertex \((0, 0)\) to \((3, -2)\).
Report an error
Skills to practice
  • Translating Graphs
I.2 Stretches
Recappreviously studied — click to reread
Definition — Vertical Stretch
A vertical stretch stretches or compresses the graph of a function vertically. It is defined, for a number \(a \neq 0\), by:$$g(x) = a \cdot f(x)$$ This transformation maps a point \((x, y)\) to \((x, ay)\).
  • If \(|a|>1\), the graph is stretched vertically by a factor of \(|a|\).
  • If \(0 < |a| < 1\), the graph is compressed vertically by a factor of \(|a|\).
  • If \(a < 0\), the stretch or compression is followed by a reflection in the \(x\)-axis (\(a = -1\) gives the reflection alone).
Figure
Definition — Horizontal Stretch
A horizontal stretch stretches or compresses the graph of a function horizontally. It is defined, for a number \(b \neq 0\), by:$$g(x) = f(bx)$$ This transformation maps a point \((x, y)\) to a new point \((\frac{x}{b}, y)\).
  • If \(|b|>1\), the graph is compressed horizontally by a factor of \(\frac{1}{|b|}\).
  • If \(0 < |b| < 1\), the graph is stretched horizontally by a factor of \(\frac{1}{|b|}\).
  • If \(b < 0\), the stretch or compression is followed by a reflection in the \(y\)-axis (\(b = -1\) gives the reflection alone).
Figure
Example
From the graph of \(f(x) = x^2\), plot \(a(x) = 2f(x)\) and \(b(x) = f(2x)\).
  • \(a(x) = 2f(x) = 2x^2\): every output is doubled — a vertical stretch with scale factor \(2\). The point \((1, 1)\) moves to \((1, 2)\).
  • \(b(x) = f(2x) = 4x^2\): the input is consumed twice as fast — a horizontal stretch with scale factor \(\frac12\), a compression: every \(x\)-coordinate is halved. The point \((1, 1)\) moves to \(\left(\frac12, 1\right)\).
The two results differ: doubling the output is not the same as doubling the input.
Figure
Stretches and Scale Factors
The number by which a stretch multiplies the distances to the axis that stays fixed is its scale factor. For \(a > 0\), the graph of \(y = a f(x)\) is the image of the graph of \(y = f(x)\) under the vertical stretch with scale factor \(a\): each point \((x, y)\) goes to \((x, ay)\). For \(b > 0\), the graph of \(y = f(bx)\) is its image under the horizontal stretch with scale factor \(\dfrac{1}{b}\): each point \((x, y)\) goes to \(\left(\dfrac{x}{b}, y\right)\). A scale factor greater than \(1\) moves the points away from the axis that stays fixed (the \(x\)-axis for a vertical stretch, the \(y\)-axis for a horizontal one), a scale factor between \(0\) and \(1\) brings them closer to it: the graph of \(y = f(2x)\) is the image of the graph of \(f\) under the horizontal stretch with scale factor \(\dfrac{1}{2}\), a compression.
Report an error
Skills to practice
  • Stretching Graphs
I.3 Reflections
Recappreviously studied — click to reread
Definition — Reflection in the x-axis
A reflection in the x-axis flips the graph of a function vertically. It is a special case of vertical stretch where \(a=-1\). It is defined by:$$g(x) = -f(x)$$ This transformation maps a point \((x, y)\) to \((x, -y)\).
Figure
Definition — Reflection in the y-axis
A reflection in the y-axis flips the graph of a function horizontally. It is a special case of horizontal stretch where \(b=-1\). It is defined by:$$g(x) = f(-x)$$ This transformation maps a point \((x, y)\) to \((-x, y)\).
Figure
Example
From the graph of \(f(x) = \sqrt{x}\), plot \(r(x) = -f(x)\) and \(s(x) = f(-x)\).
  • \(r(x) = -\sqrt{x}\): every output changes sign — a reflection in the \(x\)-axis. The curve flips below the axis; the point \((4, 2)\) moves to \((4, -2)\).
  • \(s(x) = \sqrt{-x}\): every input changes sign — a reflection in the \(y\)-axis. The curve flips to the left half-plane (defined for \(x \le 0\)); the point \((4, 2)\) moves to \((-4, 2)\).
Figure
Skills to practice
  • Reflecting Graphs
II Composite Transformations
II.1 Combining Transformations
Recappreviously studied — click to reread
When several transformations are applied to a function, the order can matter between two horizontal transformations, or between two vertical ones (a stretch and a translation, for example); a horizontal transformation and a vertical one can be applied in either order. The standard form for a transformed function is:$$ g(x) = a \cdot f(b(x-c)) + d $$ To graph this function from the parent function \(f(x)\), apply the transformations in the following order:
Report an error
Method — Order of Transformations
  1. Horizontal Transformations (inside the brackets):
    • Apply the horizontal stretch with scale factor \(\frac{1}{|b|}\) (\(b \neq 0\)), followed by a reflection in the \(y\)-axis if \(b < 0\).
    • Apply the horizontal translation (shift) by \(c\) units.
  2. Vertical Transformations (outside the brackets):
    • Apply the vertical stretch with scale factor \(|a|\), followed by a reflection in the \(x\)-axis if \(a < 0\).
    • Apply the vertical translation (shift) by \(d\) units.
Note: Always take out the coefficient \(b\) as a factor of the term inside the function to correctly identify the horizontal shift \(c\). For example, transform \(f(2x-6)\) as \(f(2(x-3))\). This shows the horizontal stretch with scale factor \(\frac{1}{2}\) (a compression) followed by a translation of 3 units to the right, not 6.
Example
Describe the sequence of transformations that maps the graph of \(f(x)=\sqrt{x}\) onto the graph of \(g(x) = 3\sqrt{-x+2} - 4\).

First, we rewrite the function \(g(x)\) in the standard form \(a \cdot f(b(x-c)) + d\):$$ g(x) = 3\sqrt{-(x-2)} - 4 $$ Comparing this to \(f(x)=\sqrt{x}\), we have \(a=3\), \(b=-1\), \(c=2\), and \(d=-4\). The transformations are applied as follows:
  1. Horizontal Transformations on \(f(x)=\sqrt{x}\):
    • Reflection: Since \(b=-1\), reflect the graph in the y-axis (\(y = \sqrt{-x}\)).
    • Translation: Since \(c=2\), translate the graph 2 units to the right (\(y = \sqrt{-(x-2)}\)).
  2. Vertical Transformations on the result:
    • Stretch: Since \(a=3\), stretch the graph vertically with scale factor 3 (\(y = 3\sqrt{-(x-2)}\)).
    • Translation: Since \(d=-4\), translate the graph 4 units down (\(y = 3\sqrt{-(x-2)} - 4\)).
Figure

Example
The graph of \(y = f(x)\) is reflected in the \(y\)-axis, then stretched horizontally with scale factor \(2\). Find an equation of the image, and check it with a point \((a, b)\) of the graph of \(f\).

Each transformation acts on the equation obtained at the previous step: a horizontal transformation replaces \(x\) in the current expression, a vertical transformation acts on the whole current expression.
  1. Reflection in the \(y\)-axis: \(x\) is replaced by \(-x\): \(y = f(-x)\).
  2. Horizontal stretch with scale factor \(2\): \(x\) is replaced by \(\dfrac{x}{2}\) in the current expression \(f(-x)\): \(y = f\left(-\dfrac{x}{2}\right)\).
The image is the graph of \(y = f\left(-\dfrac{x}{2}\right)\).
Check with a point. The reflection sends a point \((a, b)\) of the graph of \(f\) to \((-a, b)\), and the stretch doubles the \(x\)-coordinate: \((-a, b)\) goes to \((-2a, b)\). At \(x = -2a\), the expression inside \(f\) is $$\begin{aligned}[t] -\frac{x}{2} &= -\frac{-2a}{2} && (x = -2a) \\ &= a, \end{aligned}$$ so \(f\left(-\dfrac{x}{2}\right) = f(a) = b\), since the point \((a, b)\) is on the graph of \(f\): the point \((-2a, b)\) is indeed on the image. Replacing \(x\) by \(-2x\) instead, that is writing \(f(-2x)\), would give the horizontal stretch with scale factor \(\dfrac{1}{2}\): the check with a point detects this error, since \(f\big(-2 \times (-2a)\big) = f(4a)\) is not equal to \(b\) in general.
Figure

Skills to practice
  • Sketching Combined Transformations
  • Finding Equations after a Sequence of Transformations
II.2 Rational Functions as Images of \(y \equal \frac{1}{x}\)
A function of the form \(y = \dfrac{ax + b}{cx + d}\), with \(c \neq 0\) and \(ad - bc \neq 0\), can be rewritten as \(y = p + \dfrac{q}{x - h}\) by writing the numerator around the denominator. Its graph is then the image of the graph of \(y = \dfrac{1}{x}\) under a sequence of translations, stretches and reflections, and its asymptotes are the images of the two axes.
Report an error
Example
Describe a sequence of transformations that maps the graph of \(y = \dfrac{1}{x}\) onto the graph of \(y = \dfrac{2x + 1}{x - 1}\), and deduce the asymptotes of the second graph.

Write the numerator around the denominator: \(2x + 1 = 2(x - 1) + 3\). With \(f(x) = \dfrac{1}{x}\), for \(x \neq 1\): $$\begin{aligned}[t] \frac{2x + 1}{x - 1} &= \frac{2(x - 1) + 3}{x - 1} \\ &= 2 + \frac{3}{x - 1} && \text{(split the fraction)} \\ &= 3f(x - 1) + 2. \end{aligned}$$ This is the standard form \(a \cdot f(b(x - c)) + d\) with \(a = 3\), \(b = 1\), \(c = 1\) and \(d = 2\). In the order of transformations:
  1. translate \(1\) unit to the right: \(y = \dfrac{1}{x - 1}\);
  2. stretch vertically with scale factor \(3\): \(y = \dfrac{3}{x - 1}\);
  3. translate \(2\) units up: \(y = \dfrac{3}{x - 1} + 2\).
The asymptotes \(x = 0\) and \(y = 0\) of \(y = \dfrac{1}{x}\) follow the graph: the translation \(1\) unit to the right moves them to \(x = 1\) and \(y = 0\); the vertical stretch leaves these two lines in place; the translation \(2\) units up moves \(y = 0\) to \(y = 2\). The graph of \(y = \dfrac{2x + 1}{x - 1}\) has the asymptotes \(x = 1\) and \(y = 2\). The point \((1, 1)\) of \(y = \dfrac{1}{x}\) goes to \((2, 1)\), then \((2, 3)\), then \((2, 5)\); check: \(\dfrac{2 \times 2 + 1}{2 - 1} = 5\).
Figure

Skills to practice
  • Relating Rational Functions to \(y \equal \frac{1}{x}\)
III The Graph of \(y \equal \frac{1}{f(x)}\)
III.1 Reciprocal Graphs

Let \(f(x) = x^2 - 4\). Whenever \(f(x) \neq 0\), the number \(f(x)\) has a reciprocal \(\dfrac{1}{f(x)}\): how does the graph of \(y = \dfrac{1}{f(x)}\) compare with the graph of \(f\)?
  1. Compute \(f(x)\) and \(\dfrac{1}{f(x)}\) for \(x = 0\), \(1\), \(3\) and \(4\). Why are the values the same for \(x = -1\), \(-3\) and \(-4\)?
  2. Why can \(\dfrac{1}{f(x)}\) not be computed for \(x = 2\) and for \(x = -2\)? What happens to \(\dfrac{1}{f(x)}\) when \(x\) gets closer and closer to \(2\)?
  3. What happens to \(\dfrac{1}{f(x)}\) when \(x\) becomes large?
  4. For which values of \(x\) are the two numbers \(f(x)\) and \(\dfrac{1}{f(x)}\) equal?

  1. The values:
    \(x\) \(0\) \(1\) \(3\) \(4\)
    \(f(x) = x^2 - 4\) \(-4\) \(-3\) \(5\) \(12\)
    \(\dfrac{1}{f(x)}\) \(-\dfrac{1}{4}\) \(-\dfrac{1}{3}\) \(\dfrac{1}{5}\) \(\dfrac{1}{12}\)
    Since \((-x)^2 = x^2\), \(f(-x) = f(x)\): the values at \(-1\), \(-3\) and \(-4\) are those at \(1\), \(3\) and \(4\). Each reciprocal has the same sign as \(f(x)\), and the larger \(|f(x)|\) is, the smaller \(\left|\dfrac{1}{f(x)}\right|\) is.
  2. \(f(2) = f(-2) = 0\), and \(0\) has no reciprocal. Close to \(2\), \(f(x)\) is close to \(0\): \(f(2.1) = 0.41\), \(f(2.01) = 0.0401\) and \(f(1.99) = -0.0399\), so \(\dfrac{1}{f(2.1)} \approx 2.44\), \(\dfrac{1}{f(2.01)} \approx 24.9\) and \(\dfrac{1}{f(1.99)} \approx -25.1\). The absolute values \(\left|\dfrac{1}{f(x)}\right|\) become as large as desired: the line \(x = 2\) is a vertical asymptote of the graph of \(y = \dfrac{1}{f(x)}\), and so is the line \(x = -2\).
  3. When \(x\) becomes large, \(f(x) = x^2 - 4\) becomes as large as desired, so \(\dfrac{1}{f(x)}\) gets as close to \(0\) as desired: the \(x\)-axis is a horizontal asymptote.
  4. For \(f(x) \neq 0\): $$\begin{aligned}[t] f(x) = \frac{1}{f(x)} &\iff f(x)^2 = 1 && \text{(multiply by } f(x) \neq 0\text{)} \\ &\iff f(x) = 1 \ \text{ or } \ f(x) = -1 \\ &\iff x^2 = 5 \ \text{ or } \ x^2 = 3 && (f(x) = x^2 - 4) \\ &\iff x = \pm\sqrt{5} \ \text{ or } \ x = \pm\sqrt{3}. \end{aligned}$$ The two graphs meet at four points, on the lines \(y = 1\) and \(y = -1\).
Figure
The graph of \(y = \dfrac{1}{f(x)}\) has three branches, separated by the vertical asymptotes \(x = -2\) and \(x = 2\); the minimum point \((0, -4)\) of \(f\) has become the highest point \(\left(0, -\dfrac{1}{4}\right)\) of the middle branch. The Propositions “Sign, Invariant Points and Variations” and “Asymptotes of the Graph of \(y = \dfrac{1}{f(x)}\)” turn each of these observations into a rule.


Definition — The Reciprocal of a Function
Let \(f\) be a function. The reciprocal of \(f\) is the function \(\dfrac{1}{f}\) given by $$\frac{1}{f}(x) = \frac{1}{f(x)}.$$ It is defined for every \(x\) of the domain of \(f\) such that \(f(x) \neq 0\). Its graph is the graph of the equation \(y = \dfrac{1}{f(x)}\). A point lying on both the graph of \(f\) and the graph of \(y = \dfrac{1}{f(x)}\) is an invariant point.
The reciprocal \(\dfrac{1}{f}\) must not be confused with the inverse function \(f^{-1}\), which undoes \(f\): for \(f(x) = 2x\), \(\dfrac{1}{f(x)} = \dfrac{1}{2x}\), whereas \(f^{-1}(x) = \dfrac{x}{2}\).
Proposition — Sign, Invariant Points and Variations
Let \(f\) be a function.
  • Sign: wherever \(f(x) \neq 0\), the number \(\dfrac{1}{f(x)}\) has the same sign as \(f(x)\). The graphs of \(y = f(x)\) and \(y = \dfrac{1}{f(x)}\) lie on the same side of the \(x\)-axis.
  • Invariant points: a point lies on both graphs exactly when it is a point of the graph of \(f\) at which \(f(x) = 1\) or \(f(x) = -1\). The two graphs meet only on the lines \(y = 1\) and \(y = -1\).
  • Variations: on an interval where \(f(x)\) is never \(0\) and keeps the same sign, if \(f\) is increasing then \(\dfrac{1}{f}\) is decreasing, and if \(f\) is decreasing then \(\dfrac{1}{f}\) is increasing.
  • Maximum and minimum: on such an interval, if \(f(a)\) is the maximum of \(f\), then \(\dfrac{1}{f(a)}\) is the minimum of \(\dfrac{1}{f}\), and if \(f(a)\) is the minimum of \(f\), then \(\dfrac{1}{f(a)}\) is the maximum of \(\dfrac{1}{f}\). So, when the graph of \(f\) has no break near \(a\) (then \(f(x)\) keeps the sign of \(m\) for \(x\) close to \(a\)), a local maximum point \((a, m)\) of the graph of \(f\), with \(m \neq 0\), becomes a local minimum point \(\left(a, \dfrac{1}{m}\right)\) of the graph of \(y = \dfrac{1}{f(x)}\), and a local minimum point becomes a local maximum point.

  • Sign. \(f(x) \times \dfrac{1}{f(x)} = 1 > 0\): the two numbers are nonzero and have the same sign.
  • Invariant points. The point with \(x\)-coordinate \(x\) lies on both graphs when \(f(x) = \dfrac{1}{f(x)}\), with \(f(x) \neq 0\): $$\begin{aligned}[t] f(x) = \frac{1}{f(x)} &\iff f(x)^2 = 1 && \text{(multiply by } f(x) \neq 0\text{)} \\ &\iff f(x) = 1 \ \text{ or } \ f(x) = -1. \end{aligned}$$
  • Variations. Let \(u \leq v\) be two numbers of an interval \(I\) on which \(f(x)\) is never \(0\) and keeps the same sign. Then \(f(u) f(v) > 0\), and $$\begin{aligned}[t] \frac{1}{f(u)} - \frac{1}{f(v)} &= \frac{f(v) - f(u)}{f(u) f(v)}. && \text{(common denominator)} \end{aligned}$$ If \(f\) is increasing, \(f(v) - f(u) \geq 0\), so \(\dfrac{1}{f(u)} \geq \dfrac{1}{f(v)}\): \(\dfrac{1}{f}\) is decreasing on \(I\). If \(f\) is decreasing, \(f(v) - f(u) \leq 0\) and \(\dfrac{1}{f}\) is increasing on \(I\).
  • Maximum and minimum. If \(f(a)\) is the maximum of \(f\) on \(I\), then \(f(x) \leq f(a)\) for every \(x\) of \(I\), and \(f(x) f(a) > 0\), so $$\frac{1}{f(x)} - \frac{1}{f(a)} = \frac{f(a) - f(x)}{f(x) f(a)} \geq 0:$$ \(\dfrac{1}{f(a)}\) is the minimum of \(\dfrac{1}{f}\) on \(I\). The case of a minimum is the same with the inequalities reversed. When the graph of \(f\) has no break at \(a\) and \(f(a) = m \neq 0\), \(f(x)\) keeps the sign of \(m\) on an interval around \(a\), on which this applies: a local maximum point of the graph of \(f\) becomes a local minimum point, and conversely.

Proposition — Asymptotes of the Graph of \(y \equal \frac{1}{f(x)}\)
Let \(f\) be a function whose graph has no break, except at its vertical asymptotes.
  • Zeros become vertical asymptotes: if \(f(a) = 0\), the graph of \(f\) has no break at \(a\), and \(f(x) \neq 0\) for the other values of \(x\) close to \(a\), the line \(x = a\) is a vertical asymptote of the graph of \(y = \dfrac{1}{f(x)}\).
  • Large values become values close to \(0\): where \(|f(x)|\) becomes as large as desired, \(\dfrac{1}{f(x)}\) gets as close to \(0\) as desired. So if \(|f(x)|\) becomes as large as desired when \(x\) becomes large, the \(x\)-axis is a horizontal asymptote of the graph of \(y = \dfrac{1}{f(x)}\); and if the line \(x = p\) is a vertical asymptote of the graph of \(f\), the graph of \(y = \dfrac{1}{f(x)}\) comes as close as desired to the point \((p, 0)\), which does not belong to it, since a reciprocal \(\dfrac{1}{f(x)}\) is never \(0\): this point is drawn as an open circle.
  • Horizontal asymptotes: if the line \(y = q\), with \(q \neq 0\), is a horizontal asymptote of the graph of \(f\), the line \(y = \dfrac{1}{q}\) is a horizontal asymptote of the graph of \(y = \dfrac{1}{f(x)}\).

  • Zeros. Since the graph of \(f\) has no break at \(a\) and \(f(a) = 0\), \(f(x)\) gets as close to \(0\) as desired when \(x\) gets close to \(a\), without being \(0\) for \(x \neq a\). Then \(\left|\dfrac{1}{f(x)}\right| = \dfrac{1}{|f(x)|}\) becomes as large as desired: by the definition of a vertical asymptote, the line \(x = a\) is one.
  • Large values. If \(|f(x)|\) is larger than a number \(M > 0\), then \(\left|\dfrac{1}{f(x)}\right| = \dfrac{1}{|f(x)|}\) is smaller than \(\dfrac{1}{M}\), which is as small as desired when \(M\) is large enough.
  • Horizontal asymptotes. Suppose \(f(x)\) gets as close to \(q \neq 0\) as desired when \(x\) becomes large. Once \(f(x)\) is within \(\dfrac{|q|}{2}\) of \(q\), \(|f(x)| \geq \dfrac{|q|}{2}\), and $$\left|\frac{1}{f(x)} - \frac{1}{q}\right| = \frac{|q - f(x)|}{|q| \, |f(x)|} \leq \frac{2}{q^2}\,|q - f(x)|,$$ which gets as close to \(0\) as desired: \(\dfrac{1}{f(x)}\) gets as close to \(\dfrac{1}{q}\) as desired, and the line \(y = \dfrac{1}{q}\) is a horizontal asymptote.

Method — Sketching the Graph of \(y \equal \frac{1}{f(x)}\) from the Graph of \(y \equal f(x)\)
  1. Draw the vertical asymptotes: the lines \(x = a\) at the zeros \(a\) of \(f\) (each zero \(a\) such that \(f(x) \neq 0\) for the other values of \(x\) close to \(a\)).
  2. Mark the invariant points, where the graph of \(f\) meets the lines \(y = 1\) and \(y = -1\).
  3. Turn each turning point \((a, m)\) of \(f\), with \(m \neq 0\) and the graph of \(f\) without a break near \(a\), into \(\left(a, \dfrac{1}{m}\right)\), a maximum becoming a minimum and a minimum a maximum; mark the \(y\)-intercept \(\left(0, \dfrac{1}{f(0)}\right)\) when \(f(0) \neq 0\).
  4. Where \(|f(x)|\) becomes large, draw \(\dfrac{1}{f(x)}\) close to \(0\): the \(x\)-axis as a horizontal asymptote, or an open circle at \((p, 0)\) for a vertical asymptote \(x = p\) of \(f\). A horizontal asymptote \(y = q \neq 0\) of \(f\) gives the horizontal asymptote \(y = \dfrac{1}{q}\).
  5. Join the points on each interval between the asymptotes, leaving out the open circles, keeping the sign of \(f\) and reversing its variations: where \(f\) increases, \(\dfrac{1}{f}\) decreases, and where \(f\) decreases, \(\dfrac{1}{f}\) increases.
  6. The range of \(\dfrac{1}{f}\) is made of the numbers \(\dfrac{1}{y}\), for the nonzero values \(y\) taken by \(f\).
Example
Sketch the graph of \(y = \dfrac{1}{2x - 4}\) from the graph of \(f(x) = 2x - 4\).

  1. Vertical asymptote. \(2x - 4 = 0\) for \(x = 2\): the line \(x = 2\).
  2. Invariant points. \(2x - 4 = 1\) for \(x = \dfrac{5}{2}\), and \(2x - 4 = -1\) for \(x = \dfrac{3}{2}\): the points \(\left(\dfrac{5}{2}, 1\right)\) and \(\left(\dfrac{3}{2}, -1\right)\).
  3. Intercept. \(\dfrac{1}{f(0)} = \dfrac{1}{-4} = -\dfrac{1}{4}\): the point \(\left(0, -\dfrac{1}{4}\right)\). A line that is not horizontal has no turning point.
  4. Large values. \(|2x - 4|\) becomes as large as desired when \(x\) becomes large: the \(x\)-axis is a horizontal asymptote.
  5. Sign and variations. \(f\) is increasing, negative for \(x < 2\) and positive for \(x > 2\): \(\dfrac{1}{f}\) is decreasing on each of the two intervals. The left branch lies below the \(x\)-axis and falls along \(x = 2\); the right branch lies above it and comes down from the top along \(x = 2\).
Figure

Example
The graph of a quadratic function \(f\) is shown: it cuts the \(x\)-axis at \(-1\) and \(3\), and its vertex is \((1, -4)\). Sketch the graph of \(y = \dfrac{1}{f(x)}\).
Figure

  1. Vertical asymptotes. The zeros of \(f\) are \(-1\) and \(3\): the lines \(x = -1\) and \(x = 3\).
  2. Turning point. The vertex \((1, -4)\) is the minimum point of \(f\), with \(-4 \neq 0\): it becomes the local maximum point \(\left(1, -\dfrac{1}{4}\right)\).
  3. Invariant points. The parabola meets each of the lines \(y = 1\) and \(y = -1\) at two points, which lie on both graphs. With the formula \(f(x) = (x + 1)(x - 3)\), found from the intercepts and the vertex (the coefficient \(a\) of \(f(x) = a(x + 1)(x - 3)\) satisfies \(f(1) = -4a = -4\), so \(a = 1\)), their \(x\)-coordinates are \(1 \pm \sqrt{5}\) and \(1 \pm \sqrt{3}\). The \(y\)-intercept \(f(0) = -3\) gives the point \(\left(0, -\dfrac{1}{3}\right)\).
  4. Large values. \(f(x)\) becomes as large as desired when \(x\) becomes large: the \(x\)-axis is a horizontal asymptote.
  5. Sign and variations. \(f(x) > 0\) for \(x < -1\) and for \(x > 3\), \(f(x) < 0\) between \(-1\) and \(3\). The outer branches lie above the \(x\)-axis and climb along the asymptotes; the middle branch lies below it, rises from the bottom along \(x = -1\) to its highest point \(\left(1, -\dfrac{1}{4}\right)\), then falls along \(x = 3\).
Figure

Example
The graph of \(f(x) = \dfrac{2x - 2}{x + 1}\) is shown, with its asymptotes \(x = -1\) and \(y = 2\); it passes through \((1, 0)\) and \((0, -2)\). Give the domain of \(\dfrac{1}{f}\) and sketch the graph of \(y = \dfrac{1}{f(x)}\).
Figure

  1. Domain. \(\dfrac{1}{f(x)}\) needs \(f(x)\) to be defined, so \(x \neq -1\), and nonzero, so \(x \neq 1\): the domain is \(\mathbb{R} \setminus \{-1, 1\}\).
  2. Asymptotes. The zero \(1\) of \(f\) gives the vertical asymptote \(x = 1\). The horizontal asymptote \(y = 2\) of \(f\) gives the horizontal asymptote \(y = \dfrac{1}{2}\). The vertical asymptote \(x = -1\) of \(f\) gives the open circle \((-1, 0)\): the graph comes as close as desired to this point, which does not belong to it.
  3. Points. \(f(x) = 1\) when \(2x - 2 = x + 1\), that is \(x = 3\); \(f(x) = -1\) when \(2x - 2 = -x - 1\), that is \(x = \dfrac{1}{3}\): the invariant points are \((3, 1)\) and \(\left(\dfrac{1}{3}, -1\right)\). The \(y\)-intercept is \(\dfrac{1}{f(0)} = -\dfrac{1}{2}\).
  4. Check. For \(x \neq \pm 1\), \(\dfrac{1}{f(x)} = \dfrac{x + 1}{2x - 2}\), a rational function whose asymptotes are \(x = 1\) and \(y = \dfrac{1}{2}\), as found; the graph of \(y = \dfrac{1}{f(x)}\) is the graph of this rational function without its point \((-1, 0)\), where \(f\) is not defined.
Figure

Example
Let \(f(x) = x^3 - 4x + 1\). Use technology to find the vertical asymptotes and the turning points of the graph of \(y = \dfrac{1}{f(x)}\), to \(3\) significant figures, and sketch it.

Enter \(Y_1 = x^3 - 4x + 1\) in the window \(-3 \le x \le 3\), \(-5 \le y \le 6\), and read its key features with the analysis tools (Method “Reading the Key Features of a Graph on the Screen”).
  1. Vertical asymptotes. The tool zero displays \(X = -2.1149075\), \(X = 0.25410169\) and \(X = 1.8608059\): the vertical asymptotes are \(x \approx -2.11\), \(x \approx 0.254\) and \(x \approx 1.86\).
  2. Turning points. The tool maximum displays \(X = -1.1547005\), \(Y = 4.0792014\), and the tool minimum \(X = 1.1547005\), \(Y = -2.0792014\). The local maximum of \(f\) becomes a local minimum of \(\dfrac{1}{f}\), with value \(\dfrac{1}{4.0792014} \approx 0.245\), and the local minimum becomes a local maximum, with value \(\dfrac{1}{-2.0792014} \approx -0.481\): the turning points are \((-1.15, 0.245)\) and \((1.15, -0.481)\).
  3. Check on the screen. With \(Y_2 = \dfrac{1}{Y_1}\), the tools minimum and maximum applied to \(Y_2\) give the same two points. The calculator may join two branches across an asymptote with an almost vertical segment, which is not part of the graph.
Since \(|f(x)|\) becomes as large as desired when \(x\) becomes large, the \(x\)-axis is a horizontal asymptote; the four branches keep the sign of \(f\).
Figure

Skills to practice
  • Recognising Features of the Graph of \(y \equal \frac{1}{f(x)}\)
  • Finding Asymptotes and Invariant Points
  • Studying \(y \equal \frac{1}{f(x)}\) with Technology
  • Sketching the Graph of \(y \equal \frac{1}{f(x)}\)
IV Going further
Skills to practice
  • Reasoning about Transformations
  • Solving Problems with Transformations
  • Reasoning about Reciprocal Graphs
  • Solving Problems with Reciprocal Graphs
  • Quiz