Model

Model
Mathematical Model

An equation or a system of equations representing real-world phenomena. Models also represent patterns found in graphs and/or data. Usually models are not exact matches the objects or behavior they represent. A good model should capture the essential character of whatever is being modeled.

Key Formula

y = f(x)

Where:

$y$ = The output (dependent variable) — the quantity you want to predict or describe
$x$ = The input (independent variable) — the quantity you control or observe
$f$ = The function rule that defines the relationship — could be linear, quadratic, exponential, etc.

Worked Example

Problem: A coffee shop tracks the number of iced coffees sold on days with different high temperatures. At 20°C they sold 40 cups, and at 35°C they sold 100 cups. Create a linear model relating temperature to cups sold, then predict sales at 30°C.

Step 1: Choose the model type. The data suggests a linear relationship, so use the form:

y = mx + b

Step 2: Find the slope using the two data points (20, 40) and (35, 100).

m = \frac{100 - 40}{35 - 20} = \frac{60}{15} = 4

Step 3: Find the y-intercept by substituting one point into the equation. Using (20, 40):

40 = 4(20) + b \implies b = 40 - 80 = -40

Step 4: Write the complete model:

y = 4x - 40

Step 5: Use the model to predict sales at 30°C by substituting x = 30:

y = 4(30) - 40 = 120 - 40 = 80

Answer: The linear model is y = 4x − 40, which predicts 80 iced coffees sold when the temperature is 30°C.

Another Example

This example uses an exponential model instead of a linear one, showing that you must choose the model type that fits the behavior of the data — constant growth rate (linear) versus constant percentage growth (exponential).

Problem: A bacteria colony starts with 500 cells and doubles every 3 hours. Create an exponential model for the population P after t hours, then find the population after 12 hours.

Step 1: Since the population doubles at a fixed interval, choose an exponential model of the form:

P(t) = P_0 \cdot 2^{t/d}

Step 2: Identify the parameters: the initial population is 500 and the doubling time is 3 hours.

P_0 = 500, \quad d = 3

Step 3: Write the complete model:

P(t) = 500 \cdot 2^{t/3}

Step 4: Predict the population at t = 12 hours:

P(12) = 500 \cdot 2^{12/3} = 500 \cdot 2^4 = 500 \cdot 16 = 8{,}000

Answer: The exponential model is P(t) = 500 · 2^(t/3), predicting 8,000 bacteria after 12 hours.

Frequently Asked Questions

What is the difference between a mathematical model and a formula?

A formula is a fixed equation that gives an exact result, like the area of a circle A = πr². A mathematical model is an equation chosen or built to approximate real-world behavior, and it may not be perfectly accurate. All models are expressed using formulas, but not all formulas are models — a model specifically represents some external phenomenon.

How do you choose the right type of mathematical model?

Look at the shape of your data. If the data follows a straight-line trend, use a linear model. If it curves upward or downward with increasing steepness, try a quadratic or exponential model. Plotting the data on a scatter plot first is the best way to see which shape fits. You can also use residuals or regression tools to compare how well different models match the data.

Why are mathematical models not exact?

Real-world systems are influenced by many factors — randomness, measurement error, and variables you haven't accounted for. A model deliberately simplifies reality by focusing on the most important relationships. This trade-off between simplicity and accuracy is a core feature of modeling, not a flaw.

Linear Model vs. Exponential Model

	Linear Model	Exponential Model
General form	y = mx + b	y = a · bˣ
Growth pattern	Constant amount added per unit of x	Constant percentage (multiplicative) change per unit of x
Graph shape	Straight line	Curve that increases (or decreases) rapidly
Typical use	Steady, predictable change (e.g., hourly wages, distance at constant speed)	Population growth, radioactive decay, compound interest
Long-term behavior	Increases or decreases without bound at a steady rate	Grows explosively (b > 1) or decays toward zero (0 < b < 1)

Why It Matters

Mathematical modeling appears throughout algebra, statistics, and science courses whenever you translate a real situation into an equation you can analyze. In standardized tests like the SAT and ACT, you are often given data or a scenario and asked to write, interpret, or evaluate a model. Beyond school, modeling is the foundation of fields like engineering, economics, epidemiology, and climate science — any discipline where predictions must be made from observed data.

Common Mistakes

Mistake: Assuming the model is perfectly accurate and using it to make predictions far outside the range of the original data (extrapolation).

Correction: Models are approximations that work best within the range of data used to build them. A linear model for coffee sales between 20°C and 35°C may give nonsensical predictions at 0°C or 60°C. Always state the domain over which the model is reasonable.

Mistake: Using a linear model for data that clearly curves, or forcing an exponential model when the data is roughly straight.

Correction: Before writing an equation, plot the data and examine its shape. Check residuals: if a linear model's residuals show a curved pattern, a nonlinear model is likely a better fit. Let the data guide your choice of model, not the other way around.

Related Terms

Equation — The basic building block of any model
Simultaneous Equations — Models with multiple equations solved together
Graph of an Equation or Inequality — Visual representation of a model
Regression Equation — A model fitted to data using statistical methods
Least-Squares Regression Line — The best-fit linear model minimizing squared errors
Linear Equation — The simplest and most common model type
Exponential Growth — A nonlinear model for multiplicative change
Quadratic Equation — A model for parabolic or projectile-type behavior