2.8 · AHL · HL onlyPaper 1Paper 2

Transformations of graphs

In a minute

A , or stretch turns a graph you know into a new one. Changes outside the function act on yy as they look; changes inside the bracket act on xx the opposite way, so f(x−3)f(x - 3) moves right. In models, each transformation has a meaning: a delay, a new starting level, a faster cycle.

You'll be able to

  • Translate a graph with y=f(x−a)+by = f(x - a) + b, and write the translation as a vector.
  • Reflect a graph in the xx-axis with y=−f(x)y = -f(x) and in the yy-axis with y=f(−x)y = f(-x).
  • Stretch a graph with y=pf(x)y = pf(x) and y=f(qx)y = f(qx), and say which axis stays .
  • Combine transformations in the right order, and track a point, an or a domain through them.
  • Read the constants of a model as transformations of a base graph, and say what each one means in context.
Before you start2 quick questions on what this builds on
Check yourself
Write down the equation of the horizontal asymptote of the graph of y=6(0.5)x+3y = 6(0.5)^{x} + 3.
Check yourself
Find the coordinates of the vertex of the graph of y=−2x2+10x+1y = -2x^{2} + 10x + 1.

Revise first: 2.4 Key features of graphs and 2.5 Modelling with functions

Translations

A ball is thrown and its height is h(t)h(t) metres after tt seconds. A second ball follows the same path, but is thrown 22 seconds later. At time tt it has only been flying for t−2t - 2 seconds, so its height is h(t−2)h(t - 2).

So t−2t - 2 moves the graph 22 units right, not left. Throwing from 1.51.5 m higher gives h(t)+1.5h(t) + 1.5: a change outside acts on the output, as it looks.

Each of these moves is a : the graph slides without turning or changing size.

Learn this
y=f(x−a)+b  :  translation by (ab), so a units right and b units upy = f(x - a) + b \;:\; \text{translation by } \begin{pmatrix} a \\ b \end{pmatrix} \text{, so } a \text{ units right and } b \text{ units up}

In the vector, the top number is the horizontal move (positive is right) and the bottom number is the vertical move (positive is up). So (−41)\begin{pmatrix} -4 \\ 1 \end{pmatrix} moves a graph 44 units left and 11 unit up, giving y=f(x+4)+1y = f(x + 4) + 1.

Worked exampleThe graph of y=2xy = 2^{x} is translated by the vector (−35)\begin{pmatrix} -3 \\ 5 \end{pmatrix}. Write down the equation of the new graph, and the equation of its horizontal asymptote.

Try it first, then reveal one step at a time.

  1. The vector gives a=−3a = -3 (left 33) and b=5b = 5 (up 55).
0 0.5 1 1.5 2 2.5 3 3.5 0 2 4 6 8 10 12 Time, t (s) Height (m) y = h(t) y = g(t) (1.29, 9.5) (2.09, 11.5) 0.8 right, 2 up
g(t)=h(t−0.8)+2g(t) = h(t - 0.8) + 2: every point of hh moves 0.80.8 right and 22 up.
Worked exampleA ball's height is h(t)=−4.9t2+12.6t+1.4h(t) = -4.9t^{2} + 12.6t + 1.4 metres, tt seconds after it is thrown. A second ball is thrown 0.80.8 seconds later, from a point 22 m higher, and follows the same shape of path. (a) Write g(t)g(t), the second ball's height, in terms of hh. (b) Use your GDC to find the maximum point of y=h(t)y = h(t). Hence find the coordinates of the maximum point of y=g(t)y = g(t), and interpret them. Give values to 3 s.f.

Try it first, then reveal one step at a time.

  1. (a) 0.80.8 s later is a move right: replace tt by t−0.8t - 0.8. 22 m higher is a move up. So g(t)=h(t−0.8)+2g(t) = h(t - 0.8) + 2, the translation by (0.82)\begin{pmatrix} 0.8 \\ 2 \end{pmatrix}.
Check yourself
The point (6,−1)(6, -1) lies on the graph of y=f(x)y = f(x). The graph is translated by the vector (−43)\begin{pmatrix} -4 \\ 3 \end{pmatrix}. Find the image of the point.
Check yourself
Town A's population, in thousands, tt years from now is modelled by P(t)=120(1.05)tP(t) = 120(1.05)^{t}. Town B's population is modelled by Q(t)=P(t+6)Q(t) = P(t + 6). Which statement is true?

Reflections in the axes

Put a minus sign outside: y=−f(x)y = -f(x). Every output changes sign, so (x,y)(x, y) moves to (x,−y)(x, -y). The graph flips over the xx-axis.

Put the minus sign on the input: y=f(−x)y = f(-x). At xx the new graph does what the old one did at −x-x. So (x,y)(x, y) moves to (−x,y)(-x, y), and the graph flips over the yy-axis.

Each flip is a in an axis.

Learn this
y=−f(x)  :  reflection in the x-axisy=f(−x)  :  reflection in the y-axisy = -f(x) \;:\; \text{reflection in the } x\text{-axis} \qquad y = f(-x) \;:\; \text{reflection in the } y\text{-axis}

Points on the mirror line don't move: they are . So y=−f(x)y = -f(x) keeps every xx-intercept, and y=f(−x)y = f(-x) keeps the yy-intercept.

Worked exampleLet f(x)=x3−4x2+3x+5f(x) = x^{3} - 4x^{2} + 3x + 5. The graph of y=g(x)y = g(x) is the reflection of the graph of y=f(x)y = f(x) in the yy-axis. Find g(x)g(x) in expanded form.

Try it first, then reveal one step at a time.

  1. A reflection in the yy-axis is g(x)=f(−x)g(x) = f(-x). Replace every xx by (−x)(-x), keeping the brackets.
0 5 10 15 20 -6 -4 -2 0 2 4 6 Distance from the bank, x (m) Height or depth (m) (10, −5) (10, 5) riverbed y = f(x) depth y = −f(x) invariant: (0, 0) and (20, 0)
Depth d(x)=−f(x)d(x) = -f(x) is the riverbed reflected in the xx-axis. The banks, on the mirror line, stay put.
Worked exampleAcross a river 2020 m wide, the height of the riverbed relative to the water surface is f(x)=0.05x2−xf(x) = 0.05x^{2} - x metres, xx metres from one bank. Negative values are below the surface. The depth of water is d(x)=−f(x)d(x) = -f(x). Find the coordinates of the maximum point of y=d(x)y = d(x) and interpret them.

Try it first, then reveal one step at a time.

  1. y=−f(x)y = -f(x) is a reflection in the xx-axis: (x,y)(x, y) moves to (x,−y)(x, -y). The lowest point of ff becomes the highest point of dd.
Check yourself
Find the equation of the reflection of the graph of y=3x+2y = 3^{x} + 2 in the xx-axis.
Check yourself
The graph of y=f(x)y = f(x) has a local maximum point at (−2,7)(-2, 7). What is the corresponding point on the graph of y=f(−x)y = f(-x)?

Stretches

To change a model's output from metres to centimetres, multiply it by 100100: y=100f(x)y = 100f(x). Points move away from the xx-axis; points on it stay put.

This is a .

A scale factor between 00 and 11 squashes the graph towards the xx-axis, but it is still called a stretch. A negative pp also reflects: y=−2f(x)y = -2f(x) is a stretch with scale factor 22 and a reflection in the xx-axis.

Now change the input. A wheel that turns twice as fast reaches, after tt minutes, the place the old wheel reached after 2t2t minutes. Its height is h(2t)h(2t), and each event happens in half the time.

This is a , and its scale factor is the reciprocal of the number inside.

Learn this
y=pf(x)  :  vertical stretch, scale factor py=f(qx)  :  horizontal stretch, scale factor 1qy = pf(x) \;:\; \text{vertical stretch, scale factor } p \qquad y = f(qx) \;:\; \text{horizontal stretch, scale factor } \tfrac{1}{q}

Exam questions may say which axis stays fixed. A vertical stretch acts parallel to the yy-axis and leaves the xx-axis invariant. A horizontal stretch acts parallel to the xx-axis and leaves the yy-axis invariant.

Worked exampleA company's monthly profit is P(x)=−0.5x2+28x−150P(x) = -0.5x^{2} + 28x - 150 thousand dollars, when it sells xx hundred units. It reports profit in euros, using 11 dollar =0.92= 0.92 euros, so the profit in thousand euros is E(x)=0.92P(x)E(x) = 0.92P(x). (a) Describe the transformation from y=P(x)y = P(x) to y=E(x)y = E(x). (b) Find the maximum profit in thousand euros. (c) Explain why the break-even sales don't change.

Try it first, then reveal one step at a time.

  1. (a) A vertical stretch with scale factor 0.920.92, with the xx-axis invariant.
0 5 10 15 20 25 30 0 5 10 15 20 25 30 Time, t (minutes) Height (m) (15, 30) (10, 30) y = h(t) y = h(1.5t)
y=h(1.5t)y = h(1.5t) squeezes y=h(t)y = h(t) towards the vertical axis with scale factor 11.5\frac{1}{1.5}. Heights don't change; times do.
Worked exampleA rider's height on an observation wheel is h(t)=16−14cos⁡(12t)h(t) = 16 - 14\cos(12t) metres, tt minutes after boarding, with the angle in degrees. On a busy day the wheel turns 1.51.5 times as fast, and the height is H(t)=h(1.5t)H(t) = h(1.5t). (a) Describe the transformation from y=h(t)y = h(t) to y=H(t)y = H(t). (b) Find when a rider first reaches the top on the faster wheel.

Try it first, then reveal one step at a time.

  1. (a) h(qt)h(qt) with q=1.5q = 1.5: a horizontal stretch with scale factor 11.5=23\frac{1}{1.5} = \frac{2}{3}, with the vertical axis invariant.
Check yourself
Which transformation maps the graph of y=f(x)y = f(x) onto the graph of y=f(2.5x)y = f(2.5x)?
Check yourself
The graph of y=f(x)y = f(x) passes through (0,4)(0, 4), (−2,0)(-2, 0) and (3,1)(3, 1). Which of these points is invariant under the vertical stretch y=3f(x)y = 3f(x)?

Combining transformations

Start with y=x2y = x^{2}. Stretch vertically with scale factor 22, giving y=2x2y = 2x^{2}. Then translate by (0−3)\begin{pmatrix} 0 \\ -3 \end{pmatrix}, giving y=2x2−3y = 2x^{2} - 3.

Swap the order. Translating first gives y=x2−3y = x^{2} - 3. The stretch then doubles every output, the −3-3 included: y=2x2−6y = 2x^{2} - 6. That is a different graph, so order matters.

Inside the bracket the order is reversed. For y=f(qx−c)y = f(qx - c) with q>0q > 0, translate cc units right first, then stretch horizontally with scale factor 1q\frac{1}{q}. Or stretch first and then translate cq\frac{c}{q} units right, since qx−c=q(x−cq)qx - c = q\left(x - \frac{c}{q}\right).

Horizontal and vertical changes don't interfere, so the inside changes can go before, after or between the outside ones.

Try it below. Choose the base x2x^{2} and set the vertical stretch to 22 and the shift up to −3-3. The vertex sits at (0,−3)(0, -3). The sliders use their own letters, so go by their names.

Transform a graphDrag the sliders
−6−5−4−3−2−1123456−5−4−3−2−112345dashed: y = x²solid: image

y = x²

1
1
0
0

(1, 1) maps to (1.00, 1.00)

Dashed is the base curve, solid is the image. The open dot on the base lands on the filled dot.
Worked exampleLet g(x)=3f(2x−10)+1g(x) = 3f(2x - 10) + 1. (a) Describe a sequence of transformations that maps the graph of y=f(x)y = f(x) onto the graph of y=g(x)y = g(x). (b) The point (4,−2)(4, -2) lies on the graph of y=f(x)y = f(x). Find its image on the graph of y=g(x)y = g(x).

Try it first, then reveal one step at a time.

  1. (a) Inside: 2x−102x - 10. Translate by (100)\begin{pmatrix} 10 \\ 0 \end{pmatrix}, then stretch horizontally with scale factor 12\frac{1}{2}.
Worked exampleWeekly downloads of a game, in thousands, are modelled by R(t)=−0.4t3+4.5t2+3t+20R(t) = -0.4t^{3} + 4.5t^{2} + 3t + 20, for 0≤t≤120 \le t \le 12 weeks after launch. A sequel launches 55 weeks after the game, and its weekly downloads are modelled by S(t)=1.4R(t−5)S(t) = 1.4R(t - 5), with tt still counted from the first game's launch. (a) Describe the transformations from y=R(t)y = R(t) to y=S(t)y = S(t). (b) Use your GDC to find the maximum point of y=R(t)y = R(t). Hence find the maximum point of y=S(t)y = S(t) and interpret it. Give values to 3 s.f.

Try it first, then reveal one step at a time.

  1. (a) A translation by (50)\begin{pmatrix} 5 \\ 0 \end{pmatrix} and a vertical stretch with scale factor 1.41.4. One is horizontal and one vertical, so either order works.
Check yourself
The graph of y=2xy = 2^{x} is translated by (04)\begin{pmatrix} 0 \\ 4 \end{pmatrix} and then stretched vertically with scale factor 0.50.5. Find the equation of the new graph.
Check yourself
The graph of y=f(x)y = f(x) has a local minimum point at (−3,8)(-3, 8). Find the xx-coordinate of the corresponding point on the graph of y=f(0.5x+2)−1y = f(0.5x + 2) - 1.

Transformations in models

Most models in this topic are a simple graph, transformed: lines, , exponentials, cubics and waves. Each constant is one transformation, with a meaning in context.

Take a model y=asin⁡(bx)+dy = a\sin(bx) + d, with a,b>0a, b > 0 and xx in degrees. From y=sin⁡xy = \sin x: a vertical stretch with scale factor aa sets the amplitude. A horizontal stretch with scale factor 1b\frac{1}{b} sets the period, 360∘b\frac{360^{\circ}}{b}. A translation by (0d)\begin{pmatrix} 0 \\ d \end{pmatrix} sets the principal axis.

Worked exampleThe temperature in a greenhouse is modelled by T(t)=6sin⁡(15t)+18T(t) = 6\sin(15t) + 18 degrees Celsius, tt hours after 9 am, with the angle in degrees. Describe a sequence of transformations that maps the graph of y=sin⁡ty = \sin t onto the graph of y=T(t)y = T(t), and say what each one means.

Try it first, then reveal one step at a time.

  1. Vertical stretch, scale factor 66: the temperature swings 6∘C6^{\circ}\mathrm{C} above and below its middle value.

An exponential model y=kat+cy = k a^{t} + c, with k>0k > 0, works the same way. From y=aty = a^{t}: a vertical stretch with scale factor kk, then a translation by (0c)\begin{pmatrix} 0 \\ c \end{pmatrix}. The y=0y = 0 moves to y=cy = c. When 0<a<10 < a < 1, that is the level the quantity approaches over time.

Worked exampleA cup of tea cools in a room at 21∘C21^{\circ}\mathrm{C}. Its temperature is T(t)=68(0.9)t+21T(t) = 68(0.9)^{t} + 21 degrees Celsius, tt minutes after it is poured. A second cup is poured 33 minutes later, in a room at 25∘C25^{\circ}\mathrm{C}, starting 68∘C68^{\circ}\mathrm{C} above the room. Its temperature is U(t)=T(t−3)+4U(t) = T(t - 3) + 4, with tt counted from the first pour. (a) Write down the horizontal asymptote of y=U(t)y = U(t) and interpret it. (b) Use your GDC to find when the second cup reaches 50∘C50^{\circ}\mathrm{C}. Give your answer to 3 s.f.

Try it first, then reveal one step at a time.

  1. (a) UU is TT translated by (34)\begin{pmatrix} 3 \\ 4 \end{pmatrix}. The asymptote T=21T = 21 moves up 44 to U=25U = 25: the tea cools towards the room's temperature, 25∘C25^{\circ}\mathrm{C}.
Check yourself
A wave model is y=1.2sin⁡(40x)+3y = 1.2\sin(40x) + 3, with xx in degrees. It comes from y=sin⁡xy = \sin x by stretches and a translation. Find the scale factor of the horizontal stretch.
Check yourself
A drink's temperature is modelled by T(t)=55(0.85)t+19T(t) = 55(0.85)^{t} + 19 degrees Celsius, tt minutes after it is made in a room at 19∘C19^{\circ}\mathrm{C}. The model comes from y=(0.85)ty = (0.85)^{t} by a vertical stretch with scale factor 5555, then a translation by (019)\begin{pmatrix} 0 \\ 19 \end{pmatrix}. What does the 5555 represent?

Common mistakes

  • Moving y=f(x−4)y = f(x - 4) left, or y=f(x+4)y = f(x + 4) right. Inside the bracket the sign works the other way: x−4x - 4 moves the graph 44 units right.
  • Giving qq as the scale factor of y=f(qx)y = f(qx). The scale factor is 1q\frac{1}{q}, so y=f(2x)y = f(2x) halves every xx-coordinate.
  • Translating before stretching. y=pf(x)+by = pf(x) + b means stretch by pp, then translate by bb; the other order gives pf(x)+pbpf(x) + pb.
  • Confusing −f(x)-f(x) with f(−x)f(-x). −f(x)-f(x) changes the outputs and mirrors in the xx-axis; f(−x)f(-x) changes the inputs and mirrors in the yy-axis.
  • Moving the curve but not its asymptotes or domain. Transform them too: a model for 0≤t≤120 \le t \le 12 translated 55 right holds for 5≤t≤175 \le t \le 17.
  • Rounding a turning point before transforming it. Keep the GDC's full values and round only the final coordinates.

Exam tips

  • ‘Describe’ a transformation: name it and give every detail, such as ‘horizontal stretch, scale factor 12\frac{1}{2}’ or ‘translation by (−30)\begin{pmatrix} -3 \\ 0 \end{pmatrix}’. For several, give them in an order that works: the right transformations in the wrong order can lose a mark.
  • ‘Sketch’ a transformed graph: move the endpoints, turning points, intercepts and asymptotes first, then join them with the right shape. Mark schemes check where those points land.
  • When a question asks for g(x)g(x) after a transformation, writing it in terms of ff, such as g(x)=f(x−a)+bg(x) = f(x - a) + b with the numbers in, is usually enough. Substitute the formula for ff only if the question needs it.

Key takeaways

Learn thisy=f(x−a)+b  :  translation by (ab), so a units right and b units upy = f(x - a) + b \;:\; \text{translation by } \begin{pmatrix} a \\ b \end{pmatrix} \text{, so } a \text{ units right and } b \text{ units up}
Learn thisy=−f(x)  :  reflection in the x-axisy=f(−x)  :  reflection in the y-axisy = -f(x) \;:\; \text{reflection in the } x\text{-axis} \qquad y = f(-x) \;:\; \text{reflection in the } y\text{-axis}
Learn thisy=pf(x), p>0  :  vertical stretch, scale factor py=f(qx), q>0  :  horizontal stretch, scale factor 1qy = pf(x),\ p > 0 \;:\; \text{vertical stretch, scale factor } p \qquad y = f(qx),\ q > 0 \;:\; \text{horizontal stretch, scale factor } \tfrac{1}{q}
Learn thisOutside the bracket acts on yy as it looks. Inside acts on xx the opposite way: x−ax - a moves right, and qxqx divides xx-coordinates by qq.
Learn thisA vertical stretch y=pf(x)y = pf(x) keeps the xx-axis invariant; a horizontal stretch y=f(qx)y = f(qx) keeps the yy-axis invariant.
Learn thisFor y=pf(x)+by = pf(x) + b with p>0p > 0: stretch first, then translate by (0b)\begin{pmatrix} 0 \\ b \end{pmatrix}. Horizontal and vertical changes can be done in either order relative to each other.
Learn thisImage of (x0,y0)(x_{0}, y_{0}) on y=pf(qx−c)+by = pf(qx - c) + b, with q≠0q \ne 0: solve qx−c=x0qx - c = x_{0} for xx; the new yy is py0+bpy_{0} + b.
Learn thisAsymptotes, turning points and the ends of the domain move with the graph. In a model, say what each transformation means.

Practise this

11 past-paper questions on Transformations of graphs in the AI HL bank, each with a mark scheme and a worked explanation.

Practise Transformations of graphs →