Simple Linear Regression Model Study Pack

Kibin's free study pack on Simple Linear Regression Model includes a 7-section study guide, 25 quiz questions, 30 flashcards, and 5 open-ended Explain review questions. Sign up free to track your progress toward mastery, plus upload your own notes and recordings to create personalized study packs organized by course.

Last updated May 28, 2026

Topic mastery0%

Simple Linear Regression Model Study Guide

Break down simple linear regression from the equation ŷ = b₀ + b₁x to least squares estimation, residual analysis, r², and hypothesis testing on the slope — everything you need to model, validate, and interpret a linear relationship between two variables.

Key Takeaways

  • Simple linear regression models the relationship between one explanatory variable and one continuous response variable using the equation ŷ = b₀ + b₁x, where b₀ is the y-intercept and b₁ is the slope.
  • The least squares method determines b₀ and b₁ by minimizing the sum of squared residuals — the vertical distances between observed data points and the fitted regression line.
  • The slope b₁ quantifies the average change in the response variable for each one-unit increase in the explanatory variable, while b₀ gives the predicted value of y when x equals zero.
  • The coefficient of determination (r²) measures the proportion of total variability in the response variable that is explained by the linear relationship with the explanatory variable.
  • Residual analysis — checking that residuals are roughly normally distributed, have constant variance, and show no pattern — is essential for validating regression assumptions before interpreting results.
  • A hypothesis test on the slope (H₀: β₁ = 0) determines whether the linear relationship between x and y is statistically significant, using a t-statistic with n − 2 degrees of freedom.
  • Predictions made within the observed range of x (interpolation) are generally reliable, while extrapolation beyond that range can produce misleading or invalid estimates.

What Simple Linear Regression Does and When to Use It

Simple linear regression is a statistical technique for describing and quantifying a straight-line relationship between one numerical explanatory variable and one numerical response variable, and for using that relationship to make predictions.

Response and Explanatory Variables

  • The response variable (also called the dependent variable, labeled y) is the outcome being predicted or explained.
  • The explanatory variable (also called the independent or predictor variable, labeled x) is the variable used to explain or predict changes in y.
  • Simple linear regression involves exactly one explanatory variable; adding more predictors moves into multiple regression.

Conditions That Justify a Linear Model

  • A scatterplot of y versus x should show a roughly linear trend before fitting a regression line — applying a linear model to a curved relationship produces systematically wrong predictions.
  • Both variables must be measured on a numerical scale; regression does not apply directly to categorical response variables.
  • Observations should be independent of one another, meaning the value of y for one case should not influence the value for another.

The Regression Equation: Slope, Intercept, and Predicted Values

The regression model expresses the predicted value of the response variable as a linear function of the explanatory variable, and each coefficient in that equation carries a specific, interpretable meaning.

Structure of the Regression Equation

  • The fitted regression equation is written ŷ = b₀ + b₁x, where ŷ (pronounced 'y-hat') is the predicted value of y for a given value of x.
  • The population version of the model is written μy = β₀ + β₁x, where β₀ and β₁ are true (unknown) parameters that b₀ and b₁ estimate from sample data.

Interpreting the Slope (b₁)

  • The slope b₁ represents the average change in ŷ for each one-unit increase in x, holding all else constant.
  • A positive slope indicates that y tends to increase as x increases; a negative slope indicates the opposite.
  • The slope is the most substantively important coefficient because it directly describes the direction and rate of the linear relationship.

Interpreting the Y-Intercept (b₀)

  • The intercept b₀ is the predicted value of y when x = 0.
  • In many real-world contexts the intercept has no practical meaning because x = 0 falls outside the observed data range or is physically impossible — in those cases the intercept is a mathematical anchor for the line, not a meaningful prediction.

Making Predictions with the Equation

  • Substituting any specific value of x into ŷ = b₀ + b₁x yields a point estimate of the mean response at that x-value.
  • Interpolation — predicting within the range of x values used to build the model — is valid; extrapolation beyond that range is unreliable because the linear pattern may not continue.

Estimating Coefficients: The Least Squares Method

The regression line is not placed arbitrarily — a specific mathematical criterion called least squares uniquely determines the best-fitting line from the data.

Residuals: Measuring Prediction Error

  • A residual is the difference between an observed value and the model's predicted value: eᵢ = yᵢ − ŷᵢ.
  • Positive residuals indicate the model underestimated the observed value; negative residuals indicate overestimation.
  • Residuals are the raw material for assessing both model fit and assumption validity.

The Least Squares Criterion

  • The least squares method selects b₀ and b₁ to minimize the sum of squared residuals (SSE = Σeᵢ²), rather than the simple sum, so that positive and negative errors do not cancel each other out.
  • Squaring the residuals also penalizes large errors more heavily than small ones, making the line sensitive to outliers.

Formulas for the Least Squares Coefficients

  • The slope is computed as b₁ = Σ[(xᵢ − x̄)(yᵢ − ȳ)] / Σ[(xᵢ − x̄)²], which is the ratio of the covariance of x and y to the variance of x.
  • Once the slope is known, the intercept follows from b₀ = ȳ − b₁x̄, which guarantees that the fitted line always passes through the point (x̄, ȳ).
  • These formulas produce the unique line that minimizes SSE — no other straight line will yield a smaller total squared error for the same data set.

Measuring Model Fit: Variance Partitioning and r²

After fitting a regression line, it is important to quantify how well the model actually explains the variability in the response variable, rather than simply accepting the equation at face value.

Partitioning Total Variability

  • The total variability in y is measured by SST (total sum of squares) = Σ(yᵢ − ȳ)², which captures how spread out the y values are around their mean.
  • SST splits into two parts: SSR (regression sum of squares), the variability explained by the linear relationship with x, and SSE (error sum of squares), the variability left unexplained.
  • The relationship SST = SSR + SSE holds exactly for every simple linear regression model.

Coefficient of Determination (r²)

  • The coefficient of determination r² = SSR / SST expresses the proportion of total variability in y that is accounted for by the linear regression on x.
  • r² ranges from 0 to 1; a value of 0.85 means 85% of the variation in y is explained by the linear relationship with x, and 15% remains unexplained.
  • r² is the square of the Pearson correlation coefficient r, so the sign of r (positive or negative) tells you the direction of the relationship that r² alone cannot convey.

Relationship Between r² and Model Usefulness

  • A high r² does not automatically mean the model is correctly specified — a curved relationship can still produce a moderately high r² when fit with a straight line while violating linearity assumptions.
  • r² should always be interpreted alongside residual plots rather than in isolation.

Unlock the rest of this study guide

  • Access the full study pack
  • Track your mastery and be test-day ready
  • Upload your own notes to build personalized study guides, quizzes, flashcards, and more
Sign up free →

About this Study Pack

Created by Kibin to help students review key concepts, prepare for exams, and study more effectively. This Study Pack was checked for accuracy and curriculum alignment using authoritative educational sources. See sources below.

Sources

More in Statistics

See all topics →

Browse other courses

See all courses →
Simple Linear Regression Model Study Pack | Kibin