A graph represents a function exactly when each x-value has at most one y-value. This is tested with the vertical line test.
If any vertical line intersects the graph more than once, the graph is not a function. A circle, for example, is not the graph of y as a function of x because many vertical lines hit it twice.
Six parent functions are especially useful to recognize:
y=∣x∣: absolute value, V-shaped.
y=x2: quadratic, a parabola opening upward.
y=x3: cubic, increasing through the origin, kind of doing a wiggly motion.
y=x1: reciprocal, a rotated hyperbola with asymptotes x=0 and y=0.
y=x: square-root curve, starting at (0,0) and moving right.
Knowing these shapes helps you sketch the transformed functions covered later. The chart below includes exponential, logarithmic, and trig functions, which also appear in later units.
Transformations move or change a graph while preserving the basic shape of the parent function.
If c>0:
y=f(x)+c shifts the graph up c units.
y=f(x)−c shifts the graph down c units.
y=f(x+c) shifts the graph left c units.
y=f(x−c) shifts the graph right c units.
y=−f(x) reflects the graph over the x-axis.
y=f(−x) reflects the graph over the y-axis.
The input changes happen in the opposite direction from how they look. For example, f(x−3) shifts right 3 because the inside of the function reaches the old input value when x is 3 larger.
the starting point is (h,k). The basic domain is x≥h unless there is a reflection inside the radical. All transformations behave the same except it changes the domain/range of the radical.
The domain of a combined function is the intersection of the domains of the pieces, with any extra restrictions from the operation. For a quotient, the denominator must not be 0.
For example, if
f(x)=x2−21,g(x)=x,h(x)=x1,
then
domain of f:x=±2,domain of g:x≥0,domain of h:x=0.
Any arithmetic combination must respect the restrictions from all functions involved.
A function has an inverse that is also a function only if the original function is one-to-one, also called injective.
A function is injective if different inputs always produce different outputs. Equivalently, every output that the function actually hits comes from exactly one input. In symbols,
f(a)=f(b)⟹a=b.
Graphically, injective functions pass the horizontal line test: every horizontal line intersects the graph at most once.
A function is surjective, or onto, if every element of the target set is hit by the function. In other words, for every allowed output y in the codomain, there is at least one input x in the domain such that
f(x)=y.
Surjectivity depends on the codomain you choose. For example,
f(x)=x2
is not surjective as a function from R to R, because negative outputs are never reached. But it is surjective as a function from R to [0,∞).
A function is bijective if it is both injective and surjective. This means every output in the codomain is hit exactly once. Bijective functions have inverse functions that undo them perfectly on the stated domain and codomain.
For example,
y=x2
is a function because it passes the vertical line test, but it is not injective on R because it fails the horizontal line test. Both x=2 and x=−2 give the same output 4.
To make y=x2 injective, restrict the domain. For example, on x≥0,
f(x)=x2
is injective. If we define it as a function from [0,∞) to [0,∞), then it is also surjective, so it is bijective. Its inverse is
f−1(x)=x.
On x≤0, its inverse would instead be
f−1(x)=−x.
Domain restrictions are how we choose one branch when a relation would otherwise give more than one output.
A quick map of bijectivity, surjectivity, and injectivity is shown below:
For t>0, the outputs cover [0,∞). For t<0, the outputs cover (−∞,0]. Therefore
range=(−∞,∞).
Let f(x)=⎩⎨⎧x2−4x+1,ax+b,x+c,x<1,1≤x<4,x≥4.. Find a, b, and c so that the pieces connect at x=1 and x=4, and so that the middle piece has average rate of change 3 on [1,4].
The average rate of change of the middle piece ax+b is its slope, so
a=3.
The pieces connect at x=1, so
12−4(1)+1=3(1)+b.
This gives
−2=3+b⟹b=−5.
The pieces connect at x=4, so
3(4)−5=4+c.
Thus
7=4+c⟹49=4+c⟹c=45.
Therefore
a=3,b=−5,c=45.
Find the explicit form(s) for the relation (x2+y2)2=4(x2−y2).
Let
u=y2.
Then
(x2+u)2=4(x2−u).
Expand and collect terms:
x4+2x2u+u2=4x2−4uu2+(2x2+4)u+x4−4x2=0.
Using the quadratic formula,
u=2−(2x2+4)±(2x2+4)2−4(x4−4x2).
The discriminant simplifies:
(2x2+4)2−4(x4−4x2)=16(2x2+1).
So
u=−(x2+2)±22x2+1.
Since u=y2,
y2=−(x2+2)+22x2+1
or
y2=−(x2+2)−22x2+1.
The second expression is always negative, so it gives no real branches. The real explicit forms are
y=−(x2+2)+22x2+1
and
y=−−(x2+2)+22x2+1.
For these branches to be real,
−(x2+2)+22x2+1≥0.
This simplifies to x2≤4, so both branches have domain
[−2,2].
Find the explicit form(s) for the relation y2−2xy+x2=4x+4. State the domain of each branch.
Start with
y2−2xy+x2=4x+4.
This equation is quadratic in y. Move everything to one side:
y2−2xy+(x2−4x−4)=0.
Using the quadratic formula with
a=1,b=−2x,c=x2−4x−4,
we get
y=2−(−2x)±(−2x)2−4(1)(x2−4x−4).
Simplify the discriminant:
(−2x)2−4(x2−4x−4)=4x2−4x2+16x+16=16(x+1).
So
y=22x±16(x+1)=22x±4x+1.
Thus the explicit branches are
y=x+2x+1
and
y=x−2x+1.
Both branches require
x+1≥0,
so each branch has domain
[−1,∞).
Therefore the relation is not one function of x on most of its domain; it splits into
y=x+2x+1
and
y=x−2x+1
each with domain [−1,∞).
Let f(x)=(x−3)2+2 with domain x≥3. Find f−1(x), and state the domain and range of f−1.
Start with
y=(x−3)2+2.
Switch x and y:
x=(y−3)2+2.
Then
x−2=(y−3)2.
Since the original domain is x≥3, use the positive square-root branch:
y−3=x−2.
Therefore
f−1(x)=3+x−2.
The domain of f−1 is the range of f:
[2,∞).
The range of f−1 is the domain of f:
[3,∞).
Let f:[−2,∞)→[−4,∞) be defined by f(x)=(x+2)2−4. Determine whether f is injective, surjective, bijective, or none.
On [−2,∞), the parabola starts at its vertex and increases. Therefore it passes the horizontal line test, so it is injective.
Also,
f(−2)=−4,
and as x→∞, f(x)→∞. Therefore the range is
[−4,∞),
which matches the codomain. So f is surjective.
Since f is both injective and surjective,
f is bijective.
Simplify the difference quotient for f(x)=x−12. That is, simplify hf(x+h)−f(x).
The graph of y=f(x) has domain [−4,6] and range [−2,5]. Find the domain and range of g(x)=−3f(x+1x−2)+7.
For the domain, the input to f must lie in [−4,6]:
−4≤x+1x−2≤6,x=−1.
Solve the two inequalities separately:
x+1x−2≥−4⟹x+15x+2≥0,
so
x<−1orx≥−52.
Also,
x+1x−2≤6⟹x+1−5x−8≤0,
so
x≤−58orx>−1.
Intersecting these gives
domain=(−∞,−8/5]∪[−2/5,∞).
The range cannot be determined from only the information given. The inner function
x+1x−2
takes values in [−4,6] except it never equals 1. Since we only know the domain and range of f, we do not know whether removing the input 1 changes the outputs of f. If the graph of f still hits every value in [−2,5] away from input 1, then the range would be
[−8,13],
but that is an extra assumption. Therefore the correct conclusion from the stated information is
the range is not determined by the given data.
Find f−1(3) given that f(x)=2x+f(x)3x+1.
We want f−1(3), meaning we want the input x for which
f(x)=3.
Use the given relation:
f(x)=2x+f(x)3x+1.
Substitute f(x)=3:
3=2x+33x+1.
Then
3(2x+3)=3x+16x+9=3x+13x=−8.
So
f−1(3)=−38.
Let h(x)=4−∣x−1∣. Find the domain and range of h, and describe the transformations from y=x as clearly as possible.
For the domain, require
4−∣x−1∣≥0.
Then
∣x−1∣≤4,
so
−4≤x−1≤4.
Thus
domain=[−3,5].
The largest value occurs when ∣x−1∣=0:
h(1)=4=2.
The smallest value occurs when ∣x−1∣=4:
h(−3)=h(5)=0.
Therefore
range=[0,2].
This is not a single basic transformation of y=x. It is better viewed as two square-root pieces:
h(x)={x+3,5−x,−3≤x≤1,1≤x≤5.
The graph is symmetric about x=1, has endpoints (−3,0) and (5,0), and reaches its maximum at (1,2).
Let f(x)=2x+5 and g(x)=x−31. Find formulas and domains for (f∘g)(x) and (g∘f)(x).
First,
(f∘g)(x)=f(g(x))=2(x−31)+5=x−32+5.
Combine the expression under the radical:
x−32+5=x−32+5x−15=x−35x−13.
So
(f∘g)(x)=x−35x−13.
For the domain, require x=3 and
x−35x−13≥0.
The critical values are x=513 and x=3. A sign chart gives
domain of f∘g=(−∞,513]∪(3,∞).
Now
(g∘f)(x)=g(f(x))=2x+5−31.
For the domain, require
2x+5≥0
and also
2x+5−3=0.
The first condition gives x≥−25. The second condition gives
2x+5=3⟹2x+5=9⟹x=2.
Therefore
(g∘f)(x)=2x+5−31
with
domain of g∘f=[−25,∞)∖{2}.
Let f(x)=∣x−2∣+1. If the domain is restricted to [2,∞) and the codomain is [1,∞), determine whether f is bijective and find f−1(x). Then explain what goes wrong if the domain is all real numbers.
On the restricted domain [2,∞),
f(x)=∣x−2∣+1=x−2+1=x−1.
This function is increasing on [2,∞), so it is injective.
Also, if x≥2, then
f(x)=x−1≥1.
Every output y≥1 is reached by choosing x=y+1, which is at least 2. Therefore f is surjective onto [1,∞).
So f:[2,∞)→[1,∞) is bijective.
To find the inverse, write
y=x−1.
Then
x=y+1.
Thus
f−1(x)=x+1
with domain [1,∞) and range [2,∞).
If the domain is all real numbers, the function is not injective. For example,
f(1)=2andf(3)=2.
Since two different inputs give the same output, the function does not have an inverse function on all of R.
Let f(x)=⎩⎨⎧2x+a,x2+b,cx−1,x<1,1≤x<3,x≥3. Find a, b, and c so that f is continuous everywhere and f(0)=5.
Since f(0)=5 and 0<1, use the first piece:
f(0)=2(0)+a=a.
Thus
a=5.
For continuity at x=1, the left-hand value must match the value from the middle piece:
2(1)+a=12+b.
Substitute a=5:
2+5=1+b.
So
b=6.
For continuity at x=3, the value from the middle piece must match the value from the last piece:
32+b=3c−1.
Substitute b=6:
9+6=3c−1.
Then
16=3c,
so
c=316.
Therefore
a=5,b=6,c=316.
Suppose f is an odd function with domain [−5,5], range [−3,3], and f(2)=−1. Define g(x)=2f(x−1)−4. Find the domain and range of g, and find g(3) and g(−1).
The expression f(x−1) requires
x−1∈[−5,5].
So
−5≤x−1≤5.
Add 1 throughout:
−4≤x≤6.
Thus
domain of g=[−4,6].
Since f has range [−3,3], the expression 2f(x−1) has range
[−6,6].
Subtracting 4 gives
range of g=[−10,2].
Now
g(3)=2f(3−1)−4=2f(2)−4.
Since f(2)=−1,
g(3)=2(−1)−4=−6.
So
g(3)=−6.
Also,
g(−1)=2f(−1−1)−4=2f(−2)−4.
Because f is odd,
f(−2)=−f(2)=1.
Therefore
g(−1)=2(1)−4=−2.
So
g(−1)=−2.
Suppose f:A→B and g:B→C. Prove that if f and g are both injective, then g∘f is injective. Also prove that if g∘f is surjective onto C, then g must be surjective onto C.
First suppose f and g are both injective. To prove that g∘f is injective, start by assuming two inputs give the same output:
(g∘f)(a)=(g∘f)(b).
This means
g(f(a))=g(f(b)).
Since g is injective,
f(a)=f(b).
Since f is injective,
a=b.
Therefore
g∘f is injective.
Now suppose g∘f is surjective onto C. This means that for every c∈C, there is some a∈A such that
(g∘f)(a)=c.
Equivalently,
g(f(a))=c.
But f(a) is an element of B. So for every c∈C, we have found an element of B, namely f(a), that maps to c under g.
Therefore
g is surjective onto C.
(Bonus, Cauchy’s Functional Equation)
Consider a function Q⟶Q (basically taking rational inputs and giving rational outputs) such that f(x+y)=f(x)+f(y) (basically for any two rational numbers, this equation holds true for f(x)).
(A) Show that f(0)=0 and f(−x)=−f(x). What does this show about f(x)?
(B) Prove that f(nx)=nf(x) for all n∈Z (for all integer n).
(C) Prove that f(nx)=nf(x).
(D) Determine all such functions f(x) that satisfy Cauchy’s Functional Equation. Remember you not only need to find all such solutions, but prove that each one is a valid solution to the equation.
(E) The solution in part (D) is the only solution for the rationals, but there exist infinitely more solutions for the reals! Why can’t your proof in steps (A) - (D) extend to real numbers?
For n=0, this says f(0)=0. For negative n, use part (A):
f(nx)=f(−(∣n∣x))=−f(∣n∣x)=−∣n∣f(x)=nf(x).
Therefore
f(nx)=nf(x)for all n∈Z.
For part (C), apply part (B) to x/n:
f(x)=f(n⋅nx)=nf(nx).
Therefore
f(nx)=nf(x).
For part (D), let
c=f(1).
For any rational number r=nm,
f(r)=f(nm)=nf(m).
By part (B),
f(m)=mf(1)=mc.
So
f(r)=nmc=cr.
Thus every solution must have the form
f(x)=cx
for some rational constant c. Conversely, every function of the form f(x)=cx works because
f(x+y)=c(x+y)=cx+cy=f(x)+f(y).
For part (E), the proof works over Q because every rational number is a rational multiple of 1. It does not extend to all real numbers because not every real number can be built from 1 using only integer multiplication and division. Over R, there are many wild additive functions if no conditions like continuity, monotonicity, or boundedness are required.