AP Statistics: Chapter 7 - Linear Model from Summary Statistics

General Course and Document Information

• Student: Lila Rader
• Instructor: Mr. Wigert
• Course: AP Stats
• Date: 7 October 2025
• Subject: Chapter 7: Linear Model from Summary Statistics Example

Formulas and Descriptive Statistics for Linear Models

• Equation of the Linear Model: The standard form for a regression line is $ŷ = a + bx$.
• Intercept ($a$): The value of $a$ represents the y-intercept of the model.
• Slope ($b$): The value of $b$ represents the slope of the regression line.
• Calculating the Slope ($b$): The slope is derived from the correlation coefficient and the standard deviations of the variables using the formula:   * $b = r \frac{s_y}{s_x}$   * In this equation, $r$ is the correlation coefficient, $s_y$ is the standard deviation of the dependent variable, and $s_x$ is the standard deviation of the independent variable.
• Geometry of the Regression Line: A fundamental property of the regression line is that it always passes through the point defined by the means of the data, expressed as $(\bar{x}, \bar{y})$.   * This relationship can be expressed as: $\bar{y} = a + b\bar{x}$.

Case Study: Annual Wildfire Trends and Fire Suppression Costs

• Research Question: Has the annual number of wildfires been changing, on average? If so, how fast and in what way?
• Background Context:   * Wildfires are a major concern for multiple government agencies.   * Relevant Agencies: Bureau of Land Management, Bureau of Indian Affairs, Fish and Wildlife Service, National Park Service, and USDA Forest Service.   * National Interagency Fire Center: These agencies collaborate in this center, providing statistics via their website at www.nifc.gov.
• Financial Data:   * In $2004$, the combined total spent on fire suppression by these agencies was $\$890,233,000$.   * This $2004$ figure represented a decrease/downward trend, being nearly twice as much in the year $2002$.

Summary Statistics for Wildfire Data ($1982$–$2004$)

• Independent Variable ($x$): Years since $1982$.   * Mean ($\bar{x}$): $11.5$ years (which represents the middle of the range, roughly the year $1993.5$).   * Standard Deviation ($s_x$): $7.07$ years.
• Dependent Variable ($y$): Annual number of fires, measured in thousands.   * Mean ($\bar{y}$): $114.098$ fires (in thousands).   * Standard Deviation ($s_y$): Recorded as $28.341$ in reference and $28.342$ in the descriptive summary.
• Correlation Coefficient ($r$):   * $r = -0.862$.   * This negative value indicates a strong negative linear relationship between the years elapsed since $1982$ and the total count of wildfires.

Conditions and Assumptions for the Linear Model

• For a linear model to be considered appropriate, several assumptions and conditions must be verified:   * Quantitative Variables Condition: Both the "Year" (independent) and "# of fires" (dependent) are quantitative data sets. Status: Met.   * Linearity Condition: The scatter plot of the data must show a linear trend. Status: Met.   * Residual Plot Condition: There must be no patterns visible in the residual plot. Status: Met.   * Unusual Features: There should be nothing unusual, such as significant outliers, within the data. Status: Met.
• Conclusion: Because all conditions are met, it is appropriate to create the linear model for this dataset.