Exponential Regression
Interpretation
Learn to fit exponential models to real-world data using technology and interpret the meaning of \( a \) and \( b \) in \( y = a \cdot b^x \).
What Is Exponential Regression?
When data grows (or decays) by a constant percentage rather than a constant amount, a straight line won't fit. Exponential regression uses technology to find the best-fit curve of the form:
where \( a \) is the initial value (the \( y \)-intercept when \( x = 0 \)) and \( b \) is the growth/decay factor.
Why Exponential Models Matter
Linear vs. Exponential Growth
The Model: \( y = a \cdot b^x \)
Growth vs. Decay
| Exponential Growth | Exponential Decay | |
|---|---|---|
| Factor \( b \) | \( b > 1 \) | \( 0 < b < 1 \) |
| Rate \( r = b - 1 \) | Positive (+) | Negative (−) |
| Curve shape | Rises steeply | Falls toward zero |
| Real-world example | Investment earning 6%/yr | Medication leaving the body |
Using Technology for Regression
Calculators and software find \( a \) and \( b \) by minimizing the error between the model and data. You don't solve by hand: the tool does the heavy lifting.
What Does \( r^2 \) Tell You?
The coefficient of determination \( r^2 \) measures how well the model fits the data. It ranges from 0 to 1.
How to Interpret the Output
Example 1: Bacteria Growth
A biologist counts bacteria every hour. A calculator gives the regression: \( y = 50.2 \cdot 1.32^x \) with \( r^2 = 0.997 \).
Example 2: Car Depreciation
A car's value (in thousands of dollars) is tracked over years. The regression yields: \( y = 28.5 \cdot 0.84^x \) with \( r^2 = 0.991 \).
Example 3: Making a Prediction
Using the bacteria model \( y = 50.2 \cdot 1.32^x \), predict the count at hour 10.
Interpolation vs. Extrapolation
Interpreting in Context
The most important skill is translating math into meaningful statements. Compare these two answers:
Common Mistakes
Summary Reference Table
| Component | Symbol | Interpretation |
|---|---|---|
| Initial value | \( a \) | The \( y \)-value when \( x = 0 \) |
| Growth/decay factor | \( b \) | Multiplier per unit of \( x \) |
| Percent rate | \( r = b - 1 \) | Positive = growth, negative = decay |
| Goodness of fit | \( r^2 \) | Closer to 1 = better fit |
| Prediction | \( y = a \cdot b^x \) | Plug in \( x \) to forecast \( y \) |
Question 1 of 5
A regression model gives \( y = 120 \cdot 1.07^x \). What does the value 1.07 represent?
Question 2 of 5
In the model \( y = 500 \cdot 0.88^x \), what does 500 represent in context if \( x \) is years and \( y \) is the value in dollars?
Question 3 of 5
A regression gives \( r^2 = 0.42 \). Which interpretation is most accurate?
Question 4 of 5
A town's population is modeled by \( y = 2400 \cdot 1.03^x \), where \( x \) is years since 2010. What is the predicted population in 2020?
Question 5 of 5
If a model gives \( b = 0.75 \), which statement is correct?