United StatesPrecalculusSyllabus dot point

How do you build a sinusoidal model from a periodic context or data set, and how do you interpret it?

Topic 3.7 Sinusoidal Function Context and Data Modeling: construct a sinusoidal model from a periodic context or data, and use it to make and interpret predictions.

A focused answer to AP Precalculus Topic 3.7, covering how to build a sinusoidal model from a periodic context, how sinusoidal regression fits data, and how to interpret the amplitude, period, midline and phase in context.

Generated by Claude Opus 4.89 min answerUpdated 2026-06-04

Reviewed by: AI editorial process; not yet individually human-reviewed

Have a quick question? Jump to the Q&A page

Quick answer

To model a periodic context with a sinusoid $y = a\cos(b(x - c)) + d$ (or with sine), translate the situation's features into parameters: the amplitude $a$ is half the gap between the maximum and minimum, the midline $d$ is their average, the period sets $b = \frac{2\pi}{\text{period}}$ , and the phase $c$ is anchored to a known feature, such as where the maximum occurs for a cosine model. Each parameter has a real-world meaning: the midline is the average level, the amplitude is the size of the swing, the period is how often the cycle repeats, and the phase locates the cycle in time. When given raw data instead of features, a sinusoidal regression on a calculator finds the best-fitting parameters. Always state units and check the prediction against the known extremes.

Jump to a section

What this topic is asking
Building a model from a context
Interpreting the parameters
Fitting a model to data
Try this

What this topic is asking

The College Board (Topic 3.7) wants you to model a periodic situation with a sinusoidal function and interpret the result. You build the model from the context's features (maximum, minimum, period, and where the peak falls) or fit one to data using sinusoidal regression, then use it to predict values and explain what each parameter means in context.

Building a model from a context

Choosing cosine and anchoring at the maximum is usually cleanest, because most contexts state where the high point occurs.

Interpreting the parameters

Each parameter answers a real question. The midline is the long-run average (mean sea level, average daylight, resting value). The amplitude is how far the quantity departs from average at the extremes. The period is the natural cycle length (a day, a year, a tidal cycle). The phase locates the peak on the time axis. Reading these back into the context, with units, is half the marks on a modelling question.

Modelling a Ferris wheel

A rider's height repeats every $30$ seconds, with a maximum of $26$ m and a minimum of $2$ m, reaching the maximum first at $t = 7.5$ s. Write a cosine model and find the height at $t = 0$ .

step 1 Amplitude and midline

$|a| = \frac{26 - 2}{2} = 12$ ; midline $d = \frac{26 + 2}{2} = 14$ . Take $a = 12$ .

step 2 Find b from the period

Period $30$ s gives $b = \frac{2\pi}{30} = \frac{\pi}{15}$ .

step 3 Anchor the phase at the maximum

A cosine peaks at $x = c$ , and the first maximum is at $t = 7.5$ , so $c = 7.5$ . The model is $h(t) = 12\cos\left(\frac{\pi}{15}(t - 7.5)\right) + 14$ .

step 4 Evaluate at t = 0

$h(0) = 12\cos\left(\frac{\pi}{15}(-7.5)\right) + 14 = 12\cos\left(-\frac{\pi}{2}\right) + 14 = 12 \cdot 0 + 14 = 14$ m. At $t = 0$ the rider is at the midline height, $14$ m, which is consistent with reaching the peak a quarter period ( $7.5$ s) later.

Fitting a model to data

When given a table of points rather than clean features, use a sinusoidal regression on a graphing calculator to find $a$ , $b$ , $c$ and $d$ that best fit the data. The calculator minimizes the squared residuals, just as for linear or exponential regression in Unit 1 and Unit 2. You still interpret the output parameters in context and should check that the period it reports matches the visible cycle length in the data.

A point worth stating once is that a sinusoidal model is only appropriate when the context genuinely repeats. If a quantity grows without bound or settles to a limit, a sinusoid is the wrong family no matter how well a short stretch seems to fit. Confirming periodicity before fitting, the same discipline as the competing-model validation of Topic 2.6, keeps the model honest.

Try this

Q1. A quantity oscillates between $10$ and $40$ . What is its amplitude and midline? [1 point]

Cue. Amplitude $\frac{40 - 10}{2} = 15$ ; midline $y = \frac{40 + 10}{2} = 25$ .

Q2. A cycle repeats every $8$ hours. What is $b$ in the sinusoidal model? [1 point]

Cue. $b = \frac{2\pi}{8} = \frac{\pi}{4}$ .

Exam-style practice questions

Practice questions written in the style of College Board exam questions on this dot point, with worked answer explainers. The year tag is the paper they imitate, not the source.

AP 2023 (style)1 marksSection I, Part B (multiple choice, calculator allowed). Daily high tide is

6

m and low tide is

2

m, with one full tidal cycle every

12

hours. What is the midline of a sinusoidal tide model? (A)

y = 2

(B)

y = 4

(C)

y = 6

(D)

y = 8

Show worked answer →

The correct answer is (B), $y = 4$ .

The midline is the average of the high and low values: $\frac{6 + 2}{2} = 4$ . The amplitude would be $\frac{6 - 2}{2} = 2$ and the period $12$ hours, but the question asks only for the midline, which is $y = 4$ .

AP 2024 (style)4 marksSection II (free response, calculator allowed). The number of daylight hours in a city varies sinusoidally, with a maximum of

15

hours and a minimum of

9

hours over a period of

365

days. The maximum occurs on day

172

. (a) Find the amplitude, midline and period. (b) Write a cosine model

D(t) = a\cos(b(t - c)) + d

and use it to predict the daylight on day

172

Show worked answer →

A 4-point question on building and using a sinusoidal model.

(a) Features (2 points): amplitude $a = \frac{15 - 9}{2} = 3$ ; midline $d = \frac{15 + 9}{2} = 12$ ; period $365$ , so $b = \frac{2\pi}{365}$ .
(b) Model and prediction (2 points): the maximum at day $172$ anchors the cosine phase, $c = 172$ , giving $D(t) = 3\cos\left(\frac{2\pi}{365}(t - 172)\right) + 12$ . At $t = 172$ : $D(172) = 3\cos(0) + 12 = 3 + 12 = 15$ hours, the maximum, as expected.

Related dot points

Sources & how we know this

AP Precalculus Course and Exam Description — College Board (2023)