The idea, then worked examples solved step by step, then practice questions whose numbers change every time you try them.
The exponential function
An exponential function has the form
f(x)=abx where a=0, the base b is positive and b=1. The constant a is the initial value, f(0)=ab0=a. Each increase of 1 in x multiplies the output by b. When a>0, b>1 gives exponential growth and 0<b<1 gives exponential decay. The domain is all real numbers, and the outputs all have the sign of a: for a>0 the range is y>0.
Drag the sliders. However you set them, the graph never crosses the x-axis, never turns around and never changes how it bends.
Increasing or decreasing, concave up or down
Look at the outputs of f(x)=3⋅2x over the equal-length intervals [0,1],[1,2],…, and at the changes between them, which are the average rates of change over those intervals:
| x | 0 | 1 | 2 | 3 | 4 |
|---|
| f(x)=3⋅2x | 3 | 6 | 12 | 24 | 48 |
| average rate of change over the interval before | | 3 | 6 | 12 | 24 |
The rates are all positive, so f is increasing; and they increase, so the graph is concave up. The rates are themselves multiplied by b=2 at each step. This happens everywhere: an exponential function is always increasing or always decreasing, and it is always concave up or always concave down. So it has no point of inflection, and no relative maximum or minimum.
| b>1 | 0<b<1 |
|---|
| a>0 | increasing, concave up | decreasing, concave up |
| a<0 | decreasing, concave down | increasing, concave down |
The concavity depends only on the sign of a: multiplying by a negative a reflects the graph over the x-axis, and that turns concave up into concave down.
On a closed interval [m,n] the greatest and least values of an exponential function are at the ends, because it only goes one way. For f(x)=5(0.8)x on [−1,4]:
f is decreasing, so the greatest value is f(−1)=425 at the left end and the least is f(4)=125256 at the right end.
End behavior and the horizontal asymptote
For b>1, as x increases the factor bx grows without bound, and as x decreases it shrinks toward 0. A vertical translation by k moves everything up by k. For g(x)=3⋅2x+5:
x→∞limg(x)x→−∞limg(x)=∞=5 So the line y=5 is a horizontal asymptote: the graph gets as close to it as you like on the left but never reaches it. The range of g is y>5; the value 5 itself is not in it. Click the picture to watch g drawn from left to right, hugging its asymptote first.
Recognizing an exponential from outputs
Over equal-length input intervals the outputs of abx are proportional: each is the one before times the same factor. For abx+k it is the outputs minus k that are proportional. In the table of g:
| x | 0 | 1 | 2 | 3 |
|---|
| g(x) | 8 | 11 | 17 | 29 |
| g(x)−5 | 3 | 6 | 12 | 24 |
The ratios of g(x) are 811=811 and 1117=1117, not equal; but after taking 5 away every ratio is 2.
Warning.
The usual slips: reading 0<b<1 as "negative"; calling abx increasing whenever b>1, even with a<0; keeping the asymptote at y=0 after a vertical translation; putting k itself in the range of abx+k; and claiming a point of inflection, which an exponential function never has.
Note.
On the AP exam, write end behavior with limits, limx→−∞(abx+k)=k, and say what it means: "so y=k is a horizontal asymptote". Justify concavity through rates: "the average rates of change over equal-length intervals increase, so the graph is concave up".
Worked examples
Each example is one kind of question the AP exam asks about exponential functions. Every number in the working is found by the engine.
Example.
1. Reading the behavior from a and b. Let f(x)=−4(0.6)x. Is f increasing or decreasing? Is its graph concave up or concave down? Give its range and its end behavior.
Here a=−4<0 and 0<b=0.6<1. The factor 0.6x shrinks as x increases, so −4(0.6)x gets closer to 0 from below: f is increasing. Because a<0 the graph is concave down. The average rates of change over [0,1], [1,2], [2,3] confirm both:
| x | 0 | 1 | 2 | 3 |
|---|
| f(x) | −4 | −2.4 | −1.44 | −0.864 |
| average rate of change over the interval before | | 1.6 | 0.96 | 0.576 |
The rates are positive (increasing) and get smaller (concave down). Every output has the sign of a, so the range is y<0. End behavior:
x→∞limf(x)x→−∞limf(x)=0=−∞ The usual slip is to call f decreasing because b<1: with a<0 the reflection reverses that.
Example.
2. A shifted exponential from a table. The table gives values of a function f of the form f(x)=abx+k, and limx→−∞f(x)=3. Find a, b and k.
| x | 0 | 1 | 2 | 3 |
|---|
| f(x) | 7 | 11 | 19 | 35 |
| f(x)−3 | 4 | 8 | 16 | 32 |
The limit says the horizontal asymptote is y=3, so k=3. The outputs themselves are not proportional (711=1119), but the outputs minus 3 are: each is 2 times the one before. So b=2, and at x=0, ab0=a=4:
f(x)=4⋅2x+3 Check at x=3: 4⋅23+3=35, which is the table's value.
Example.
3. Asymptote and range after a translation. Let g(x)=−2(31)x+6. Find the end behavior of g, its horizontal asymptote and its range.
x→∞limg(x)x→−∞limg(x)=6=−∞ As x increases, (31)x shrinks toward 0, so g(x) approaches 6: the line y=6 is a horizontal asymptote, on the right this time. As x decreases, (31)x grows without bound and the factor −2 sends g(x) down without bound. Since −2(31)x is always negative, g(x)<6 for every x: the range is y<6.
Example.
4. Greatest and least values on a closed interval. A pie is taken out of an oven. Its temperature t hours later is T(t)=200(0.75)t+70 degrees Fahrenheit, for 0≤t≤6. Find the greatest and least temperatures on that interval.
Here a=200>0 and b=0.75<1, so T is decreasing on the whole interval. A function that only decreases has its greatest value at the left end and its least at the right end:
T(0)T(6)=200(0.75)0+70=270=200(0.75)6+70≈105.596 The greatest temperature is 270 degrees Fahrenheit, when the pie comes out, and the least is about 105.596 degrees Fahrenheit, after 6 hours. There is no relative minimum inside the interval: the temperature keeps falling toward the room's 70 degrees, the horizontal asymptote, without reaching it.
Example.
5. Repeated multiplication. A culture starts with 500 bacteria, and the number triples every hour. How many are there after 6 hours? Write the number after t hours as an exponential function.
Each hour multiplies the number by 3, so after 6 hours it has been multiplied by 3 six times:
N(6)=500⋅3⋅3⋅3⋅3⋅3⋅3=500⋅36=364500 After t hours, N(t)=500⋅3t: the initial value is a=500 and the base is b=3, the factor for each hour. The formula also gives values between whole hours, such as N(2.5)≈7794.229, assuming the growth goes on smoothly. Writing 500⋅3t or 500+3t treats repeated multiplication as a single product or as repeated addition.
Answer AP-style questions on exponential functions. Each answer is checked, with the full solution, and every try brings new numbers. Practice needs a free account: it keeps your progress.