Combinations of transformations
Theory
Combining transformations applies a dilation, a reflection and a translation to a base function \(y=f(x)\) in one rule, \(y=a\,f\big(n(x-h)\big)+k\). The order matters — dilations and reflections act first, translations last — so tracking a key point through the sequence and reading off the final domain, range and asymptotes pins down the transformed graph.
A combination of transformations takes a standard function \(y=f(x)\) — such as \(y=x^{2}\), \(y=\sqrt{x}\) or \(y=\dfrac{1}{x}\) — and applies a dilation, a reflection and a translation together. Every combination can be written in the single form \(y=a\,f\big(n(x-h)\big)+k\), and reading off \(a,n,h,k\) tells you exactly which transformations are involved.
Each parameter has a fixed meaning. \(a\) is a dilation by factor \(|a|\) from the \(x\)-axis (a reflection in the \(x\)-axis as well when \(a<0\)); \(n\) is a dilation by factor \(\dfrac{1}{|n|}\) from the \(y\)-axis (a reflection in the \(y\)-axis when \(n<0\)); \(h\) is a horizontal translation (right when \(h>0\)); and \(k\) is a vertical translation (up when \(k>0\)).
The order of application is not free: dilations and reflections are applied first, then the translations last. A convenient way to see the effect is to track a key point — the turning point of a parabola, the endpoint of \(y=\sqrt{x}\), or the crossing of the asymptotes of \(y=\dfrac1x\). Under the combination, a point \((x,y)\) on the base graph moves to \(\left(\dfrac{x}{n}+h,\;a\,y+k\right)\), and the image's domain, range and asymptotes follow from that key feature.
The general form of a combination of transformations of \(y=f(x)\):
Where each point on the base graph is mapped by:
For a transformed reciprocal, the asymptotes of \(y=\dfrac1x\) move to \(x=h\) and \(y=k\):
How to apply a combination of transformations
- Identify the base function and the parameters. Match the rule to \(y=a\,f\big(n(x-h)\big)+k\), or read \(a,n,h,k\) from a described sequence of a dilation, a reflection and a translation.
- Apply in the correct order. Carry out the dilations and reflections first (the \(a\) and \(n\) factors), then the translations last (the \(h\) and \(k\) shifts).
- Track a key point. Send a distinctive feature through \((x,y)\mapsto\left(\dfrac{x}{n}+h,\;a\,y+k\right)\) — the turning point, endpoint, or crossing of the asymptotes.
- State domain, range and asymptotes. Read them from the image: horizontal changes adjust the domain, vertical changes adjust the range, and translations move any asymptotes to \(x=h,\;y=k\).
| \(y\) | \(=\) | \(2x^{2}\) |
| \(y\) | \(=\) | \(2x^{2}+3\) |
| \(y\) | \(=\) | \(-3\sqrt{x}\) |
| \(y\) | \(=\) | \(-3\sqrt{x-2}+1\) |
| domain | \(:\) | \(x\ge2\) |
| range | \(:\) | \(y\le1\) |
| \(a\) | \(=\) | \(-1\) |
| \(h\) | \(=\) | \(-1\) |
| \(k\) | \(=\) | \(4\) |
| \(y\) | \(=\) | \(\dfrac{2}{x}\) |
| \(y\) | \(=\) | \(-\dfrac{2}{x}\) |
| \(y\) | \(=\) | \(-\dfrac{2}{x-3}+1\) |
| asymptotes | \(:\) | \(x=3,\;y=1\) |
| domain | \(:\) | \(x\neq3\) |
| range | \(:\) | \(y\neq1\) |
Common pitfalls
Frequently asked questions
What is the general form for a combination of transformations?
The image is \(y=a\,f\big(n(x-h)\big)+k\) from the base \(y=f(x)\): \(a\) is a dilation by \(|a|\) from the \(x\)-axis (reflection in the \(x\)-axis if \(a<0\)), \(n\) a dilation by \(\dfrac{1}{|n|}\) from the \(y\)-axis (reflection in the \(y\)-axis if \(n<0\)), \(h\) a horizontal translation and \(k\) a vertical translation.
In what order are combined transformations applied?
Dilations and reflections first (the \(a\) and \(n\) factors), then the translations last (the \(h\) and \(k\) shifts) — matching the rule \(y=a\,f\big(n(x-h)\big)+k\) read from the inside out.
How do you track a key point through several transformations?
A point \((x,y)\) maps to \(\left(\dfrac{x}{n}+h,\;a\,y+k\right)\). Apply this to a feature such as a turning point, an endpoint, or the crossing of the asymptotes to place it on the image.
How do transformations change the domain and range?
Horizontal transformations (\(n\) and \(h\)) change the domain; vertical transformations (\(a\) and \(k\)) change the range. E.g. \(y=\sqrt{x}\) becomes \(y=-2\sqrt{x-1}+3\) with domain \(x\ge1\) and range \(y\le3\).
What happens to the asymptotes of a reciprocal graph under a translation?
The asymptotes \(x=0\) and \(y=0\) of \(y=\dfrac1x\) are unmoved by dilations and reflections but shifted by a translation, so \(y=\dfrac{a}{n(x-h)}+k\) has asymptotes \(x=h\) and \(y=k\).
Does the order of transformations matter?
Yes — a dilation then a translation generally differs from the same translation then dilation, because a dilation scales the shift. Following dilations and reflections first, translations last, avoids the error.