Course contentsShow
← Data Science
Lesson 519 of 2,14514. Simple Linear RegressionFree lesson

Computing β₁: The Slope Estimate

Learn the formula for estimating the slope using covariance and variance, and interpret it as the change in y per unit x.

Computing β₁: The Slope Estimate

What you'll learn: How to calculate the slope of your regression line using a simple formula involving covariance and variance.

The Core Formula

Now that you understand why we minimize squared residuals, it's time to compute the actual slope estimate, denoted as β₁ (beta-one). The least squares method gives us this elegant formula:

β₁ = Cov(x, y) / Var(x)

In other words:

  • β₁ = covariance of x and y divided by variance of x

What This Means in Plain English

The covariance measures how x and y move together. The variance of x measures how spread out your x-values are. By dividing them, you get the average change in y for each one-unit increase in x.

Think of it like climbing a hill: if you walk forward 1 meter (change in x), how many meters up do you climb (change in y)? That ratio is your slope.

Interpreting β₁

If β₁ = 2.5, it means:

  • For every 1-unit increase in x, y increases by 2.5 units on average
  • Positive β₁ = positive relationship (upward slope)
  • Negative β₁ = negative relationship (downward slope)
  • β₁ near zero = flat line, little linear relationship

Why This Formula Works

The covariance captures the direction and strength of the linear relationship. Dividing by variance of x standardizes this to "per unit of x," giving you the rate of change. It's the mathematically optimal slope that minimizes all those squared residuals you learned about earlier.

Key Takeaway: The slope estimate β₁ equals covariance(x,y) divided by variance(x), and represents the expected change in y per one-unit increase in x.