Mathematical modeling / English

How to calculate linear regression using least square method

statisticsfun · YouTube · 8:29

YouTube · statisticsfunOriginal video

Selected moment · 0:13 · Chapter markers are time references, not playback progress.

IN THIS VIDEO

01Independent and dependent variables in the example
02Scatter plot of the five observations
03Means of x and y

FROM THE LEARNING NOTES

\bar{x}=3,\quad \bar{y}=4

Means of x and y

SELECTED MOMENT

Define independent and dependent variables

SUGGESTED PREPARATION

Basic coordinate plotting
Arithmetic mean
Signed numbers and subtraction
Arithmetic with signed numbers
Concept of the mean of a data set
Reading values from a scatter plot and table
Basic idea of a straight-line equation
Sample mean
Summation notation
Basic linear equation solving
Slope-intercept form of a line

READ & KEEP

The explanation, unpacked.

Saves 0

Reviewed learning material · Video analysis · English

Read the full overview

This 180-second excerpt introduces a simple linear-regression setup. After a brief preview of the final formulas, the narrator builds a scatter plot from made-up data with independent variable $x = 1$ , 2, 3, 4, 5 and dependent variable $y = 2$ , 4, 5, 4, 5. He computes the means $\bar{x}=3$ and $\bar{y}=4$ , marks the mean point (3,4), and states that regression lines must pass through that point. The clip then shifts from geometry to algebra by adding a column $x-\bar{x}$ and calculating the signed deviations -2, -1, 0, 1, 2 for the five observations. This 180-second clip demonstrates part of a least-squares linear regression calculation for five paired observations. It first draws the horizontal mean line $y=\bar{y}=4$ and fills the y-deviation column $y-\bar{y}$ . The video then states the target model $\hat{y}=b_0+b_1x$ and prepares the slope computation by adding two table columns: ( $x-\bar{x}$ )^2 and ( $x-\bar{x}$ )( $y-\bar{y}$ ). After summing those columns to 10 and 6, it computes $b_1=6/10=0.6$ . The intercept $b_0$ is named in the model but is not calculated within this excerpt. This 149-second clip completes a worked least-squares regression example and then previews the next topic. It first turns the table totals into the slope formula $b_1=\frac{\sum(x-\bar{x})(y-\bar{y})}{\sum(x-\bar{x})^2}$ and computes $b_1=0.6$ . It then uses the stated fact that the regression line passes through the mean point $(3,4)$ to solve $4=b_0+0.6(3)$ , obtaining $b_0=2.2$ . The final fitted line is written as $\hat{y}=2.2+0.6x$ . After that, the video only introduces $R^2$ conceptually with a distance diagram; no full $R^2$ formula is shown in this segment.

Use the learning inspector for key ideas and moments, or open the reading tabs for the complete notes.

Chapters

0:00Introduction and preview of regression results 0:13Define independent and dependent variables 0:41Plot the five observations 1:11Compute the means and mark (3,4)1:42Visualize deviations from the mean lines 2:04Build the x - x̄ column 3:00Mean line for y 3:09Computing

y - \bar{y}

3:50Regression equation goal 4:08Squared x-deviations 4:30Cross-product deviations 5:18Sums and slope

b_1

6:00Slope from the deviation table 6:18Switching to the intercept 6:30Using the mean point (3,4)7:05Solving for b0 7:46Final regression equation 8:06Preview of R-squared 8:21Outro

Learning script

Generated from the video's visuals and explanation; not verbatim speech.

The clip opens by naming the topic as regression and briefly warning that the finished formulas will look complicated. A short preview slide already shows the eventual outputs: a fitted equation, coefficient values, and a fuller table with deviation and product columns.

The narrator then rebuilds the problem from scratch. First he introduces the independent variable as x and places its values 1, 2, 3, 4, 5 in a table beside a horizontal axis. Next he introduces the dependent variable as y and adds the paired values 2, 4, 5, 4, 5 with a vertical axis. This establishes the two-variable structure needed for a scatter plot.

Each row of the tables is converted into a point on the coordinate plane. The plotted observations are (1,2), (2,4), (3,5), (4,4), and (5,5). The visual emphasis is on turning paired numerical data into a geometric picture before any line is fitted.

The next step computes the center of the data in each direction. The mean of the x-values is stated as 3, and the mean of the y-values is stated as 4. These are drawn as a vertical line at $x=3$ and a horizontal line at $y=4$ , whose intersection is labeled (3,4). The narrator states the key property that regression lines must pass through this mean point.

After locating the mean point, the video motivates the algebraic quantities used in least-squares regression. Dotted guide segments show how far each observation lies from the mean lines in the x-direction and y-direction. The spoken wording calls these offsets distances, but the later table makes clear that the recorded x-offsets are signed deviations.

The final part of the clip adds a new column headed $x-\bar{x}$ . Using $\bar{x}=3$ , the narrator subtracts the mean from each x-value in order: $1-3=-2$ , $2-3=-1$ , $3-3=0$ , $4-3=1$ , and $5-3=2$ . By the end of the segment, the example has moved from raw paired data to the first explicit deviation column needed for the least-squares calculation.

The clip opens on a scatter plot of five points together with a table that already contains x, y, and $x-\bar{x}$ . The speaker turns attention to the dependent variable and states that the mean of y is 4. Visually, a green horizontal line is drawn at $y=4$ , establishing the reference level $\bar{y}$ .

Using that mean line, the video computes the deviations $y-\bar{y}$ one row at a time. The spoken arithmetic is 2-4, 4-4, 5-4, 4-4, and 5-4, producing the column entries -2, 0, 1, 0, and 1. Dashed vertical connectors between points and the mean line reinforce that these numbers are signed vertical distances from the mean.

After the deviation table is partly complete, the speaker announces the larger objective: obtain the regression equation $\hat{y}=b_0+b_1x$ . A dashed straight line is drawn through the scatter plot as the intended fitted line, and the narration narrows the immediate task to determining the slope parameter $b_1$ .

To build the slope formula, the video first creates the column ( $x-\bar{x}$ )^2. Each existing x-deviation is squared, giving 4, 1, 0, 1, and 4. This step converts signed horizontal deviations into nonnegative contributions that will later form the denominator of $b_1$ .

Next, a second auxiliary column is added: ( $x-\bar{x}$ )( $y-\bar{y}$ ). The speaker multiplies the x-deviation and y-deviation in each row, obtaining 4, 0, 0, 0, and 2. Rows where either factor is zero contribute nothing to the numerator, which is made explicit by the repeated multiplications by zero.

With both auxiliary columns filled, the clip sums them. The squared-deviation column totals 10, and the cross-product column totals 6. The speaker identifies 10 as the denominator and 6 as the numerator, then writes $b_1=\frac{6}{10}=0.6$ . The excerpt ends with the slope determined, while the intercept $b_0$ remains part of the model but is not computed here.

The segment begins with the completed deviation table and the numeric slope computation $b_1=\frac{6}{10}=.6$ . The speaker points to the columns involving $y-\bar{y}$ and $(x-\bar{x})^2$ , then rewrites the quotient in symbolic form as $b_1=\frac{\sum(x-\bar{x})(y-\bar{y})}{\sum(x-\bar{x})^2}$ . The key mathematical move here is recognizing that the numerator is the sum of cross-products of deviations and the denominator is the sum of squared x-deviations.

Having identified $b_1$ , the lesson shifts to $b_0$ , described aloud as the y-intercept. The graph is used to motivate the next step: the fitted line is said to pass through the mean point $(3,4)$ , where $3=\bar{x}$ and $4=\bar{y}$ . This is not proved in the clip, but it is stated explicitly and visually reinforced by guide lines on the plot.

The equation $4=b_0+.6(3)$ is written by substituting the mean point and the already computed slope into the regression model $\hat{y}=b_0+b_1x$ . The arithmetic then proceeds in clear stages: first $.6(3)=1.8$ , giving $4=b_0+1.8$ ; next, subtract $1.8$ from both sides; finally simplify to $2.2=b_0$ . Red cancellation marks emphasize the elimination of the $\pm1.8$ terms.

With both coefficients known, the video consolidates the result in a summary box: $b_0=2.2$ , $b_1=.6$ , and therefore $\hat{y}=2.2+.6x$ . This is the completed least-squares regression line for the five displayed data points.

The final portion changes topic. A title card introduces “R Squared” / “R²”, and a new scatter plot labels distances such as “distance estimated - mean” and “distance actual - mean”. The speaker says the next discussion will compare distances between regression-line estimates and actual values. Within this clip, however, $R^2$ is only previewed conceptually; no formula or derivation is provided before the outro.

Knowledge cards

Independent and dependent variables in the example

The video sets up a regression example by assigning x to the independent variable and y to the dependent variable. The listed values are $x = 1$ , 2, 3, 4, 5 and $y = 2$ , 4, 5, 4, 5.

Scatter plot of the five observations

The paired data are plotted as the points (1,2), (2,4), (3,5), (4,4), and (5,5). This converts the tables into a geometric picture that will later support the regression discussion.

Means of x and y

The narrator computes the sample means directly from the listed data and states that the mean of x is 3 and the mean of y is 4. These are drawn as the vertical line $x=3$ and the horizontal line $y=4$ .

\bar{x}=3,\quad \bar{y}=4

Regression lines pass through the mean point

The clip states that all regression lines have to go through the point where the two means cross. In this example that point is (3,4), i.e. ( $\bar{x}$ , $\bar{y}$ ).

(\bar{x},\bar{y})=(3,4)

From mean offsets to the deviation column

The video first shows dotted offsets from the mean lines and then formalizes the x-offsets in a new table column $x-\bar{x}$ . Although the narration says "distance," the displayed entries are signed deviations.

x-\bar{x}

Computed x-deviations for the five rows

Using $\bar{x}=3$ , the example calculates each deviation in order: $1-3=-2$ , $2-3=-1$ , $3-3=0$ , $4-3=1$ , and $5-3=2$ . The completed column is -2, -1, 0, 1, 2.

1-3=-2,\;2-3=-1,\;3-3=0,\;4-3=1,\;5-3=2

Horizontal mean line for y

The video draws a green horizontal line at $y=4$ and identifies it as the mean of the y-values. This line serves as the reference from which vertical deviations $y-\bar{y}$ are measured.

\bar{y}=4

Computing y-deviations

Each y observation is subtracted by the mean 4. The completed column is -2, 0, 1, 0, 1, showing how far each point lies below, on, or above the mean line.

y_i-\bar{y}

Target regression model

The clip states that the goal is to find the equation of the fitted line, written as $\hat{y}=b_0+b_1x$ , and that the immediate unknown to determine is the slope $b_1$ .

\hat{y}=b_0+b_1x

Squared x-deviation column

To prepare the denominator for the slope, the table adds ( $x-\bar{x}$ )^2. The values become 4, 1, 0, 1, 4 after squaring the existing x-deviations.

(x_i-\bar{x})^2

Cross-product deviation column

The numerator for the slope is built by multiplying paired deviations row by row: ( $x-\bar{x}$ )( $y-\bar{y}$ ). The resulting entries are 4, 0, 0, 0, 2.

(x_i-\bar{x})(y_i-\bar{y})

Slope formula from column sums

The video sums the two auxiliary columns, obtaining 10 in the denominator and 6 in the numerator, then computes the least-squares slope as $b_1=6/10=0.6$ .

b_1=\frac{\sum (x_i-\bar{x})(y_i-\bar{y})}{\sum (x_i-\bar{x})^2}=\frac{6}{10}=0.6

Least-squares slope formula

The slope of the fitted line is computed from deviation sums: $b_1=\frac{\sum(x-\bar{x})(y-\bar{y})}{\sum(x-\bar{x})^2}$ . In the example, the table totals give $b_1=\frac{6}{10}=0.6$ .

b_1=\frac{\sum (x-\bar{x})(y-\bar{y})}{\sum (x-\bar{x})^2}

Regression line passes through the mean point

The video states that the fitted line must cross $(\bar{x},\bar{y})=(3,4)$ . This fact is used to substitute known values into the regression equation and solve for the intercept.

(\bar{x},\bar{y})=(3,4)

Solving for the intercept b0

Substitute the mean point and slope into $\hat{y}=b_0+b_1x$ : $4=b_0+0.6(3)$ . Simplify to $4=b_0+1.8$ , subtract 1.8 from both sides, and obtain $b_0=2.2$ .

4=b_0+.6(3)\Rightarrow b_0=2.2

Final fitted regression equation

Once $b_0=2.2$ and $b_1=0.6$ are known, the completed model is written as $\hat{y}=2.2+0.6x$ .

\hat{y}=2.2+0.6x

R-squared preview only

The clip ends by introducing $R^2$ as the next topic and showing a distance-based diagram with labels like “distance estimated - mean” and “distance actual - mean”. No full definition or formula for $R^2$ appears in this segment.

Detailed learning notes

Explore conditions, steps and evidence. Supplementary explanations are labeled separately from content shown in the video.

Symbols · 23

x

Clear evidence

Shown in the video

Evidence

Audio
Observation
The narrator says, "I'm going to start with the independent variable which is the x axis".
Diagram
Observation
A table labeled "independent variable" has column header x with values 1, 2, 3, 4, 5; a horizontal axis is labeled x.

Symbol

Meaning

Independent variable; the horizontal-axis observations used in the regression example.

Domain

Observed numeric values in this example are 1, 2, 3, 4, 5.

y

Clear evidence

Shown in the video

Evidence

Audio
Observation
The narrator says, "And also the dependent variable which is the y axis".
Diagram
Observation
A table labeled "dependent variable" has column header y with values 2, 4, 5, 4, 5; a vertical axis is labeled y.

Symbol

Meaning

Dependent variable; the vertical-axis observations paired with each x value.

Domain

Observed numeric values in this example are 2, 4, 5, 4, 5.

$\bar{x}$

Clear evidence

Shown in the video

Evidence

Audio
Observation
The narrator says, "I'm going to take the mean of the x values ... which is three".
Diagram
Observation
Below the data table, the label "mean" appears with 3 under the x column; later a new column header is written as x - x̄.

Symbol

$\bar{x}$

Meaning

Mean of the x-values in the sample.

Domain

In this example, $\bar{x}=3$ .

$\bar{y}$

Clear evidence

Shown in the video

Evidence

Audio
Observation
The narrator says, "And then I'm going to take the mean or the average of the y values ... which is four".
Diagram
Observation
Below the data table, the label "mean" appears with 4 under the y column.

Symbol

$\bar{y}$

Meaning

Mean of the y-values in the sample.

Domain

In this example, $\bar{y}=4$ .

$x-\bar{x}$

Clear evidence

Shown in the video

Evidence

Audio
Observation
The narrator says, "So I'm going to take the x value minus the mean at each point".
Diagram
Observation
A third table column is added with header x - x̄ and filled row by row with -2, -1, 0, 1, 2.

Symbol

$x-\bar{x}$

Meaning

Signed deviation of each observation's x-value from the mean of x.

Domain

For the five listed observations, the displayed values are -2, -1, 0, 1, 2.

x

Clear evidence

Shown in the video

Evidence

Formula
Observation
The table header shows the variable x.
Diagram
Observation
The scatter plot uses x as the horizontal coordinate.

Symbol

Meaning

Independent-variable observations in the data set.

Domain

Observed values shown in the table are 1, 2, 3, 4, and 5.

y

Clear evidence

Shown in the video

Evidence

Formula
Observation
The table header shows the variable y.
Diagram
Observation
The scatter plot uses y as the vertical coordinate.

Symbol

Meaning

Dependent-variable observations in the data set.

Domain

Observed values shown in the table are 2, 4, 5, 4, and 5.

$\bar{x}$

Clear evidence

Shown in the video

Evidence

Formula
Observation
The table contains the column heading $x - \bar{x}$ .
Audio
Observation
The speaker refers to the mean of x as 3 when discussing deviations.

Symbol

$\bar{x}$

Meaning

Mean of the x-values.

Domain

In this example, $\bar{x}=3$ .

$\bar{y}$

Clear evidence

Shown in the video

Evidence

Formula
Observation
The table contains the column heading $y - \bar{y}$ .
Audio
Observation
The speaker says the mean of y is 4 and draws a horizontal mean line at that value.

Symbol

$\bar{y}$

Meaning

Mean of the y-values.

Domain

In this example, $\bar{y}=4$ .

$\hat{y}$

Clear evidence

Shown in the video

Evidence

Formula
Observation
The equation written on screen is $\hat{y} = b_0 + b_1 x$ .
Audio
Observation
The speaker states the goal is to get the equation y hat equals b naught plus b one times x.

Symbol

$\hat{y}$

Meaning

Predicted or fitted value of y from the regression line.

Domain

Defined by the linear model $\hat{y}=b_0+b_1x$ .

$b_0$

Clear evidence

Shown in the video

Evidence

Formula
Observation
The symbol $b_0$ appears in the displayed equation $\hat{y} = b_0 + b_1 x$ .

Uncertainties

This clip does not compute $b_0$ explicitly.

Symbol

$b_0$

Meaning

Intercept term in the linear regression equation.

Domain

Part of the fitted model $\hat{y}=b_0+b_1x$ .

$b_1$

Clear evidence

Shown in the video

Evidence

Formula
Observation
The symbol $b_1$ appears in $\hat{y} = b_0 + b_1 x$ and later in the slope calculation.
Audio
Observation
The speaker says he is determining what b one is and later calls it the slope.

Symbol

$b_1$

Meaning

Slope coefficient of the regression line.

Domain

Computed in this clip as $b_1=0.6$ .

Knowledge points · 16

Independent variable in the regression setup

Clear evidence

Shown in the video

Evidence

Audio
Observation
The narrator introduces "the independent variable which is the x axis".
Diagram
Observation
A table titled "independent variable" lists x values 1 through 5, and a horizontal axis labeled x is drawn.

Definition

Explanation

The video defines the horizontal variable as the independent variable and assigns it to the x-axis. In the worked example, the independent-variable values are 1, 2, 3, 4, 5.

Formula

Conditions

Used for the plotted explanatory variable on the horizontal axis.

Dependent variable in the regression setup

Clear evidence

Shown in the video

Evidence

Audio
Observation
The narrator introduces "the dependent variable which is the y axis".
Diagram
Observation
A second table titled "dependent variable" lists y values 2, 4, 5, 4, 5, and a vertical axis labeled y is drawn.

Definition

Explanation

The video defines the vertical variable as the dependent variable and assigns it to the y-axis. In the worked example, the dependent-variable values are 2, 4, 5, 4, 5.

Formula

Conditions

Used for the plotted response variable on the vertical axis.

Mean of the x-values

Clear evidence

Shown in the video

Evidence

Audio
Observation
The narrator says, "I'm going to take the mean of the x values ... which is three".
Diagram
Observation
The word mean appears below the table with 3 under the x column, and a vertical line is drawn at $x=3$ .

Formula

Explanation

The example computes the mean of the five x-values and states that it equals 3. This value is then represented geometrically by a vertical line through $x=3$ .

Formula

\bar{x}=3

Conditions

Computed from the listed x-values 1, 2, 3, 4, 5.

Prerequisites

Independent variable in the regression setup

Mean of the y-values

Clear evidence

Shown in the video

Evidence

Audio
Observation
The narrator says, "And then I'm going to take the mean or the average of the y values ... which is four".
Diagram
Observation
The word mean appears below the table with 4 under the y column, and a horizontal line is drawn at $y=4$ .

Formula

Explanation

The example computes the mean of the five y-values and states that it equals 4. This value is then represented geometrically by a horizontal line through $y=4$ .

Formula

\bar{y}=4

Conditions

Computed from the listed y-values 2, 4, 5, 4, 5.

Prerequisites

Dependent variable in the regression setup

Deviation of each x-value from the mean

Clear evidence

Shown in the video

Evidence

Audio
Observation
The narrator says, "So I'm going to take the x value minus the mean at each point".
Diagram
Observation
A new column header x - x̄ is added, and entries -2, -1, 0, 1, 2 are filled in row by row.

Method

Explanation

The video constructs a new table column by subtracting the mean of x from each observed x-value. The resulting signed deviations are shown explicitly for all five observations.

Formula

x-\bar{x}

Conditions

Apply to each listed observation using $\bar{x}=3$ .

Prerequisites

Mean of the x-values

Horizontal mean line for y

Clear evidence

Shown in the video

Evidence

Audio
Observation
The speaker says, 'Now I do the same thing with the y values. I'll draw in the mean line again, which is mean of four.'
Diagram
Observation
A green horizontal line is drawn across the scatter plot at $y=4$ .

Definition

Explanation

The video constructs the mean of the dependent variable as a horizontal reference line. In this data set, the y-values have mean 4, so the line $y=\bar{y}=4$ is drawn through the scatter plot.

Formula

y=\bar{y},\quad \bar{y}=4

Conditions

The y-values in the displayed table are 2, 4, 5, 4, and 5.

Deviations of y from its mean

Clear evidence

Shown in the video

Evidence

Audio
Observation
The speaker computes 2 minus 4, then 4 minus 4, 5 minus 4, 4 minus 4, and 5 minus 4.
Formula
Observation
The table column $y - \bar{y}$ is filled with -2, 0, 1, 0, and 1.
Diagram
Observation
Dashed vertical segments connect points to the horizontal mean line while the deviations are being discussed.

Method

Explanation

Each y observation is compared with the mean $\bar{y}=4$ by subtraction. The resulting signed deviations measure how far each point lies above or below the horizontal mean line.

Formula

y_i-\bar{y}

Conditions

Use the same y-values shown in the table.
$\bar{y}=4$ in this example.

Prerequisites

Horizontal mean line for y

Linear regression target equation

Clear evidence

Shown in the video

Evidence

Audio
Observation
The speaker says the goal is to get the equation y hat equals b naught plus b one times x, and that he will determine what b one is.
Formula
Observation
The equation $\hat{y} = b_0 + b_1 x$ is written on screen.
Diagram
Observation
A dashed straight line is drawn through the scatter plot as the intended regression line.

Definition

Explanation

The clip identifies the fitted model as a straight line in the form $\hat{y}=b_0+b_1x$ , where $b_1$ is the slope to be estimated from the data and $b_0$ is the intercept.

Formula

\hat{y}=b_0+b_1x

Conditions

The relationship being modeled is linear in x.

Squared x-deviation column

Clear evidence

Shown in the video

Evidence

Audio
Observation
The speaker says, 'So we take this column x minus x bar and we square it,' then lists the squared values.
Formula
Observation
A new column ( $x - \bar{x}$ )^2 is added and filled with 4, 1, 0, 1, and 4.

Method

Explanation

To prepare the denominator for the slope formula, the video squares each x-deviation $x_i-\bar{x}$ . This removes sign and produces nonnegative contributions for each observation.

Formula

(x_i-\bar{x})^2

Conditions

The x-deviations already computed in the table are -2, -1, 0, 1, and 2.

Prerequisites

Linear regression target equation

Product of x- and y-deviations

Clear evidence

Shown in the video

Evidence

Audio
Observation
The speaker says to make another column from the x-minus-mean and y-minus-mean columns and multiply them row by row.
Formula
Observation
A new column ( $x - \bar{x}$ )( $y - \bar{y}$ ) is added and filled with 4, 0, 0, 0, and 2.

Method

Explanation

The numerator for the slope estimate is built by multiplying each x-deviation by the corresponding y-deviation. Positive products occur when both deviations share the same sign; zero occurs if either deviation is zero.

Formula

(x_i-\bar{x})(y_i-\bar{y})

Conditions

Use the paired deviations from the same observation row.

Prerequisites

Deviations of y from its mean
Squared x-deviation column

Least-squares slope from deviation sums

Clear evidence

Shown in the video

Evidence

Audio
Observation
The speaker says he adds up the squared-deviation column to get 10 as the denominator, adds up the product column to get 6 as the numerator, and concludes $b_1=0.6$ .
Formula
Observation
The sums 10 and 6 are written beneath the respective columns, followed by $b_1 = 6/10$ = .6.

Formula

Explanation

The slope of the fitted line is obtained by dividing the sum of cross-products of deviations by the sum of squared x-deviations. In this worked example, the two required sums are read directly from the table.

Formula

b_1=\frac{\sum (x_i-\bar{x})(y_i-\bar{y})}{\sum (x_i-\bar{x})^2}

Conditions

The denominator must be nonzero.
The formula is applied to the same set of paired observations.

Prerequisites

Squared x-deviation column
Product of x- and y-deviations

Linear regression equation form

Clear evidence

Shown in the video

Evidence

Formula
Observation
Top equation $\hat{y}=b_0+b_1x$ remains visible through the worked example.
Audio
Observation
The speaker identifies $b_0$ as the y-intercept and $b_1$ as the slope.

Definition

Explanation

The video uses the simple linear regression model $\hat{y}=b_0+b_1x$ , where $b_1$ is the slope and $b_0$ is the y-intercept. The final substituted equation is written as $\hat{y}=2.2+.6x$ .

Formula

\hat{y}=b_0+b_1x

Conditions

Simple linear regression with one explanatory variable x.
Parameters are estimated from a finite data set.

Claims and conditions · 6

Regression lines pass through the mean point

Clear evidence

Shown in the video

Evidence

Audio
Observation
The narrator says, "It turns out that all regression lines have to go through that point there. That's where the mean of the independent variable and the mean of the dependent variable cross."
Diagram
Observation
A dashed line is drawn through the intersection of the vertical line $x=3$ and horizontal line $y=4$ , and the point is labeled (3, 4).

Proposition

Statement

All regression lines have to go through the point where the mean of the independent variable and the mean of the dependent variable cross.

Hypotheses

The context is the linear-regression setup being demonstrated in the video.
The point referred to is ( $\bar{x}$ , $\bar{y}$ ).

Quantifiers

universal over the regression lines discussed in the video

Computed slope for the example

Clear evidence

Shown in the video

Evidence

Audio
Observation
The speaker says, 'So this is equal to point six. So b one is equal to point six, which is the slope.'
Formula
Observation
The displayed computation is $b_1 = 6/10$ = .6.

Proposition

Statement

For the displayed five-point data set, the least-squares slope is $b_1=0.6$ .

Hypotheses

The x-values are 1, 2, 3, 4, 5.
The y-values are 2, 4, 5, 4, 5.
$\bar{x}=3$ and $\bar{y}=4$ .

Quantifiers

For this specific data set shown in the clip.

Sum of squared x-deviations

Clear evidence

Shown in the video

Evidence

Audio
Observation
The speaker says he adds up the squared-deviation column and gets 10, calling it the denominator.
Formula
Observation
The number 10 is written under the ( $x - \bar{x}$ )^2 column.

Proposition

Statement

For the example table, $\sum (x_i-\bar{x})^2=10$ .

Hypotheses

The x-deviations are -2, -1, 0, 1, and 2.

Quantifiers

For the five displayed rows.

Sum of cross-products

Clear evidence

Shown in the video

Evidence

Audio
Observation
The speaker says he adds up the product column and gets 6, calling it the numerator.
Formula
Observation
The number 6 is written under the ( $x - \bar{x}$ )( $y - \bar{y}$ ) column.

Proposition

Statement

For the example table, $\sum (x_i-\bar{x})(y_i-\bar{y})=6$ .

Hypotheses

The paired deviation products are 4, 0, 0, 0, and 2.

Quantifiers

For the five displayed rows.

Regression line passes through the mean point

Clear evidence

Shown in the video

Evidence

Audio
Observation
“The regression line has to cross point three four.”
Diagram
Observation
The graph marks the intersection at $(3,4)$ , matching the displayed means $\bar{x}=3$ and $\bar{y}=4$ .

Uncertainties

The video states the fact but does not prove it.

Proposition

Statement

For this least-squares regression setup, the fitted line passes through $(\bar{x},\bar{y})=(3,4)$ .

Hypotheses

A least-squares regression line has been fitted to the data.
$\bar{x}=3$ and $\bar{y}=4$ are the sample means shown in the table.

Quantifiers

For the displayed data set and its fitted regression line.

Standard textbook form of the slope

Clear evidence

Shown in the video

Evidence

Audio
Observation
“And that would be the typical formula you’d see in your textbooks.”
Formula
Observation
The symbolic expression $b_1=\frac{\sum(x-\bar{x})(y-\bar{y})}{\sum(x-\bar{x})^2}$ is shown.

Proposition

Statement

The symbolic expression shown for $b_1$ is presented as the typical textbook formula for the least-squares slope.

Hypotheses

The data are organized as paired $(x,y)$ observations with sample means $\bar{x}$ and $\bar{y}$ .

Quantifiers

For ordinary simple linear regression as used in the example.

Derivations and proofs · 8

Computing the x-deviation column

Clear evidence

Shown in the video

Evidence

Audio
Observation
The narrator verbally computes each subtraction: 1 minus 3, 2 minus 3, 3 minus 3, 4 minus 3, and 5 minus 3.
Diagram
Observation
The x - x̄ column is filled sequentially with -2, -1, 0, 1, 2.

Numerical verification

Steps

Expression
$1-3=-2$
Explanation
For the first observation, subtract the mean of x from $x=1$ .
Justification
Direct computation shown in the table and stated aloud.
Shown in the video
Expression
$2-3=-1$
Explanation
For the second observation, subtract the mean of x from $x=2$ .
Justification
Direct computation shown in the table and stated aloud.
Shown in the video
Expression
$3-3=0$
Explanation
For the third observation, subtract the mean of x from $x=3$ .
Justification
Direct computation shown in the table and stated aloud.
Shown in the video
Expression
$4-3=1$
Explanation
For the fourth observation, subtract the mean of x from $x=4$ .
Justification
Direct computation shown in the table and stated aloud.
Shown in the video
Expression
$5-3=2$
Explanation
For the fifth observation, subtract the mean of x from $x=5$ .
Justification
Direct computation shown in the table and stated aloud.
Shown in the video

Conclusion

The completed x - x̄ column is -2, -1, 0, 1, 2.

Derivation of the y-deviation column

Clear evidence

Shown in the video

Evidence

Audio
Observation
The speaker verbally computes each y-minus-mean value in order.
Formula
Observation
The $y - \bar{y}$ column is filled row by row.

Numerical verification

Steps

Expression
$\bar{y}=4$
Explanation
The mean of the y-values is identified before subtracting it from each observation.
Justification
Stated in the audio and represented by the horizontal mean line at $y=4$ .
Shown in the video
Expression
$2-4=-2$
Explanation
The first y observation is below the mean by 2 units.
Justification
Direct subtraction shown in the table and spoken aloud.
Shown in the video
Expression
$4-4=0$
Explanation
The second y observation lies exactly on the mean line.
Justification
Direct subtraction shown in the table and spoken aloud.
Shown in the video
Expression
$5-4=1$
Explanation
The third y observation is above the mean by 1 unit.
Justification
Direct subtraction shown in the table and spoken aloud.
Shown in the video
Expression
$4-4=0$
Explanation
The fourth y observation again lies on the mean line.
Justification
Direct subtraction shown in the table and spoken aloud.
Shown in the video
Expression
$5-4=1$
Explanation
The fifth y observation is above the mean by 1 unit.
Justification
Direct subtraction shown in the table and spoken aloud.
Shown in the video

Conclusion

The completed y-deviation column is -2, 0, 1, 0, 1.

Derivation of squared x-deviations

Clear evidence

Shown in the video

Evidence

Audio
Observation
The speaker says to square all the values in the x-minus-mean column and recites the results.
Formula
Observation
The ( $x - \bar{x}$ )^2 column is populated with 4, 1, 0, 1, 4.

Numerical verification

Steps

Expression
$(-2)^2=4$
Explanation
The first x-deviation is squared.
Justification
Arithmetic shown in the new table column and stated in the audio.
Shown in the video
Expression
$(-1)^2=1$
Explanation
The second x-deviation is squared.
Justification
Arithmetic shown in the new table column and stated in the audio.
Shown in the video
Expression
$0^2=0$
Explanation
The third x-deviation is squared.
Justification
Arithmetic shown in the new table column and stated in the audio.
Shown in the video
Expression
$1^2=1$
Explanation
The fourth x-deviation is squared.
Justification
Arithmetic shown in the new table column and stated in the audio.
Shown in the video
Expression
$2^2=4$
Explanation
The fifth x-deviation is squared.
Justification
Arithmetic shown in the new table column and stated in the audio.
Shown in the video

Conclusion

The squared-deviation column is 4, 1, 0, 1, 4, whose sum is 10.

Derivation of the cross-product column

Clear evidence

Shown in the video

Evidence

Audio
Observation
The speaker multiplies the x-deviation and y-deviation row by row and reads out each product.
Formula
Observation
The ( $x - \bar{x}$ )( $y - \bar{y}$ ) column is filled with 4, 0, 0, 0, 2.

Numerical verification

Steps

Expression
$(-2)(-2)=4$
Explanation
Row 1 multiplies a negative x-deviation by a negative y-deviation.
Justification
Direct multiplication shown in the table and spoken aloud.
Shown in the video
Expression
$(-1)(0)=0$
Explanation
Row 2 has zero y-deviation, so the product is zero.
Justification
Direct multiplication shown in the table and spoken aloud.
Shown in the video
Expression
$(0)(1)=0$
Explanation
Row 3 has zero x-deviation, so the product is zero.
Justification
Direct multiplication shown in the table and spoken aloud.
Shown in the video
Expression
$(1)(0)=0$
Explanation
Row 4 has zero y-deviation, so the product is zero.
Justification
Direct multiplication shown in the table and spoken aloud.
Shown in the video
Expression
$(2)(1)=2$
Explanation
Row 5 multiplies positive deviations.
Justification
Direct multiplication shown in the table and spoken aloud.
Shown in the video

Conclusion

The cross-product column is 4, 0, 0, 0, 2, whose sum is 6.

Derivation of the slope $b_1$

Clear evidence

Shown in the video

Evidence

Audio
Observation
The speaker says the denominator is 10, the numerator is 6, and therefore $b_1=0.6$ .
Formula
Observation
The displayed fraction is $b_1 = 6/10$ = .6.

Numerical verification

Steps

Expression
$\sum (x_i-\bar{x})^2=4+1+0+1+4=10$
Explanation
The squared x-deviations are summed to form the denominator.
Justification
Column sum written on screen and stated in the audio.
Shown in the video
Expression
$\sum (x_i-\bar{x})(y_i-\bar{y})=4+0+0+0+2=6$
Explanation
The cross-products are summed to form the numerator.
Justification
Column sum written on screen and stated in the audio.
Shown in the video
Expression
$b_1=\frac{6}{10}$
Explanation
The slope is the numerator divided by the denominator.
Justification
Formula application shown on screen.
Shown in the video
Expression
$b_1=0.6$
Explanation
The fraction is simplified to a decimal.
Justification
Arithmetic simplification shown on screen and spoken aloud.
Shown in the video

Conclusion

The least-squares slope for the example is $b_1=0.6$ .

Derivation of the slope from the deviation table

Clear evidence

Shown in the video

Evidence

Formula
Observation
Table columns show $(x-\bar{x})^2$ summing to 10 and $(x-\bar{x})(y-\bar{y})$ summing to 6.
Formula
Observation
The written computation is $b_1=\frac{6}{10}=.6$ .
Audio
Observation
The speaker explains that the quotient uses the sums of those columns.

Numerical verification

Steps

Expression
$\sum (x-\bar{x})^2 = 10$
Explanation
The squared-deviation column for x totals 10.
Justification
Read directly from the table’s bottom total.
Shown in the video
Expression
$\sum (x-\bar{x})(y-\bar{y}) = 6$
Explanation
The cross-product column totals 6.
Justification
Read directly from the table’s bottom total.
Shown in the video
Expression
$b_1=\frac{6}{10}$
Explanation
Substitute the two totals into the slope formula.
Justification
Use of $b_1=\frac{\sum (x-\bar{x})(y-\bar{y})}{\sum (x-\bar{x})^2}$ .
Shown in the video
Expression
$b_1=.6$
Explanation
Simplify the fraction to obtain the slope.
Justification
Arithmetic simplification.
Shown in the video

Conclusion

The least-squares slope for the displayed data is $b_1=0.6$ .

Derivation of the intercept using the mean point

Clear evidence

Shown in the video

Evidence

Formula
Observation
The equation $4=b_0+.6(3)$ is written on screen.
Formula
Observation
Subsequent lines show $4=b_0+1.8$ , then $-1.8=b_0-1.8$ , then $2.2=b_0$ .
Audio
Observation
The speaker says the line must cross $(3,4)$ , substitutes those values, and subtracts 1.8 from both sides.

Proof

Steps

Expression
$(\bar{x},\bar{y})=(3,4)$
Explanation
The mean point is identified from the table and graph.
Justification
Displayed means and spoken statement that the regression line crosses $(3,4)$ .
Shown in the video
Expression
$4=b_0+.6(3)$
Explanation
Substitute $x=3$ , $\hat{y}=4$ , and $b_1=0.6$ into the regression equation.
Justification
Use of $\hat{y}=b_0+b_1x$ at the mean point.
Shown in the video
Expression
$4=b_0+1.8$
Explanation
Compute the product $0.6\times 3$ .
Justification
Arithmetic multiplication.
Shown in the video
Expression
$4-1.8=b_0+1.8-1.8$
Explanation
Subtract 1.8 from both sides to isolate $b_0$ .
Justification
Algebraic equivalence transformation of an equation.
Shown in the video
Expression
$2.2=b_0$
Explanation
Simplify both sides to obtain the intercept.
Justification
Arithmetic subtraction and cancellation.
Shown in the video

Conclusion

The intercept is $b_0=2.2$ .

Assembling the final regression equation

Clear evidence

Shown in the video

Evidence

Formula
Observation
Summary box lists $b_0=2.2$ , $b_1=.6$ , and $\hat{y}=2.2+.6x$ .
Audio
Observation
The speaker verbally restates both coefficients and the final equation.

Numerical verification

Steps

Expression
$b_0=2.2$
Explanation
Insert the computed intercept.
Justification
Result from the previous derivation.
Shown in the video
Expression
$b_1=.6$
Explanation
Insert the computed slope.
Justification
Result from the slope derivation.
Shown in the video
Expression
$\hat{y}=2.2+.6x$
Explanation
Substitute both coefficients into the general regression form.
Justification
Use of $\hat{y}=b_0+b_1x$ .
Shown in the video

Conclusion

The fitted line for the example is $\hat{y}=2.2+0.6x$ .

Worked examples · 3

Worked setup for a least-squares regression example

Clear evidence

Shown in the video

Evidence

Audio
Observation
The narrator says the numbers are made up, plots the points, computes the means, states that regression lines pass through (3,4), and then calculates x minus the mean for each point.
Diagram
Observation
The screen shows tables of x and y, a scatter plot of the five points, mean lines at $x=3$ and $y=4$ , the labeled point (3,4), and a new column x - x̄ with values -2, -1, 0, 1, 2.

Problem

Using a made-up dataset, set up the variables, plot the observations, locate the mean point, and begin computing deviations needed for regression.

Given

Independent variable values $x = 1$ , 2, 3, 4, 5.
Dependent variable values $y = 2$ , 4, 5, 4, 5.
The narrator states these numbers are made up.

Goal

Construct the scatter plot, compute $\bar{x}$ and $\bar{y}$ , identify the mean point, and calculate $x-\bar{x}$ for each observation.

Steps

Expression
$(1,2),(2,4),(3,5),(4,4),(5,5)$
Explanation
Plot the five paired observations on the coordinate plane.
Justification
The table values are transferred to the graph as points.
Shown in the video
Expression
$\bar{x}=3$
Explanation
Compute the mean of the x-values.
Justification
Stated directly by the narrator and shown below the table.
Shown in the video
Expression
$\bar{y}=4$
Explanation
Compute the mean of the y-values.
Justification
Stated directly by the narrator and shown below the table.
Shown in the video
Expression
$(\bar{x},\bar{y})=(3,4)$
Explanation
Mark the intersection of the mean lines as the point through which regression lines must pass.
Justification
The narrator states this property while the point is labeled on screen.
Shown in the video
Expression
$x-\bar{x}:-2,-1,0,1,2$
Explanation
Subtract the mean of x from each x-value to form the deviation column.
Justification
Each subtraction is computed aloud and entered into the table.
Shown in the video

Answer

The example establishes the scatter plot, the mean point (3,4), and the x-deviation column -2, -1, 0, 1, 2.

Verification

The displayed arithmetic matches the listed x-values and the stated mean $\bar{x}=3$ .

Worked least-squares slope example with five points

Clear evidence

Shown in the video

Evidence

Diagram
Observation
A scatter plot with five points and a table of x, y, deviations, squared deviations, and cross-products is shown throughout.
Audio
Observation
The speaker works through the y-deviations, then the squared x-deviations, then the cross-products, and finally the slope.

Problem

Given the displayed paired data, compute the quantities needed for the least-squares regression line and determine the slope $b_1$ .

Given

The table lists (x,y) = (1,2), (2,4), (3,5), (4,4), (5,5).
$\bar{x}=3$ .
$\bar{y}=4$ .

Goal

Fill the deviation columns, compute the needed sums, and find $b_1$ in $\hat{y}=b_0+b_1x$ .

Steps

Expression
$y_i-\bar{y}: -2, 0, 1, 0, 1$
Explanation
Subtract the y-mean from each y observation.
Justification
Mean-deviation calculation shown in the table and narrated.
Shown in the video
Expression
$(x_i-\bar{x})^2: 4, 1, 0, 1, 4$
Explanation
Square each x-deviation.
Justification
Preparation of the denominator for the slope formula.
Shown in the video
Expression
$(x_i-\bar{x})(y_i-\bar{y}): 4, 0, 0, 0, 2$
Explanation
Multiply each x-deviation by the corresponding y-deviation.
Justification
Preparation of the numerator for the slope formula.
Shown in the video
Expression
$\sum (x_i-\bar{x})^2=10$
Explanation
Add the squared x-deviations.
Justification
Column total written beneath the table.
Shown in the video
Expression
$\sum (x_i-\bar{x})(y_i-\bar{y})=6$
Explanation
Add the cross-products.
Justification
Column total written beneath the table.
Shown in the video
Expression
$b_1=\frac{6}{10}=0.6$
Explanation
Divide the cross-product sum by the squared-deviation sum.
Justification
Least-squares slope formula applied to the computed totals.
Shown in the video

Answer

$b_1=0.6$

Verification

The arithmetic can be checked directly from the displayed columns: $4+1+0+1+4=10$ and $4+0+0+0+2=6$ , giving $6/10=0.6$ .

Worked least-squares regression example

Clear evidence

Shown in the video

Evidence

Diagram
Observation
A five-point scatter plot and a table with rows $(1,2),(2,4),(3,5),(4,4),(5,5)$ are shown.
Formula
Observation
Means are listed as $\bar{x}=3$ and $\bar{y}=4$ ; totals are 10 and 6; final equation is $\hat{y}=2.2+.6x$ .
Audio
Observation
The speaker walks through computing $b_1$ , then $b_0$ , then the final equation.

Problem

Given five paired observations, compute the least-squares regression line $\hat{y}=b_0+b_1x$ .

Given

Data points: $(1,2),(2,4),(3,5),(4,4),(5,5)$ .
Sample means: $\bar{x}=3$ , $\bar{y}=4$ .
Deviation totals: $\sum(x-\bar{x})^2=10$ , $\sum(x-\bar{x})(y-\bar{y})=6$ .

Goal

Find $b_1$ , $b_0$ , and write the fitted regression equation.

Steps

Expression
$b_1=\frac{\sum (x-\bar{x})(y-\bar{y})}{\sum (x-\bar{x})^2}$
Explanation
Start from the standard slope formula shown in symbolic form.
Justification
Formula stated in the video.
Shown in the video
Expression
$b_1=\frac{6}{10}=.6$
Explanation
Substitute the column totals from the table.
Justification
Numerical values are read from the displayed sums.
Shown in the video
Expression
$4=b_0+.6(3)$
Explanation
Use the fact that the regression line passes through $(3,4)$ .
Justification
Spoken claim plus substitution into $\hat{y}=b_0+b_1x$ .
Shown in the video
Expression
$4=b_0+1.8$
Explanation
Multiply $0.6$ by $3$ .
Justification
Arithmetic.
Shown in the video
Expression
$2.2=b_0$
Explanation
Subtract $1.8$ from both sides to solve for the intercept.
Justification
Algebraic isolation of $b_0$ .
Shown in the video
Expression
$\hat{y}=2.2+.6x$
Explanation
Write the final fitted line using the computed coefficients.
Justification
Substitution into the regression model.
Shown in the video

Answer

$b_1=0.6$ , $b_0=2.2$ , and the regression line is $\hat{y}=2.2+0.6x$ .

Verification

The video verifies the result by displaying the summary box with $b_0=2.2$ , $b_1=.6$ , and $\hat{y}=2.2+.6x$ , and by showing the line crossing the marked mean point $(3,4)$ .

Visual events · 16

Opening title and final-result preview

Clear evidence

Shown in the video

Evidence

Diagram
Observation
A title slide reads "Regression" with "David Longstreet" and "www.MyBookSucks.Com". It then cuts to a dense preview slide showing a scatter plot, a yellow box with $b0 = 2.2$ , b1 = .6, ŷ = 2.2 + .6x, a table with columns x, y, x - x̄, y - ȳ, (x - x̄)², (x - x̄)(y - ȳ), and formulas including $b1 = 6/10$ = .6 = Σ(x - x̄)(y - ȳ)/Σ(x - x̄)².
Audio
Observation
The narrator says, "In this tutorial, I'm going to talk about regression" and warns that the final equations and formulas look nasty.

Uncertainties

The preview slide appears only briefly before the step-by-step construction begins.

Objects

Title text "Regression"
Author and website text
Scatter plot preview
Yellow result box
Data table preview
Formula preview

Changes

The video moves from a title card to a full preview of the finished regression worksheet.

Invariants

The topic remains regression throughout the opening.

Interpretation

The opening previews the destination of the lesson before the narrator rebuilds the calculation from basic data.

Construction of axes and data tables

Clear evidence

Shown in the video

Evidence

Diagram
Observation
A blank white screen gains a table labeled "independent variable" with x values 1 through 5, then a horizontal axis labeled x, then a second table labeled "dependent variable" with y values 2, 4, 5, 4, 5, then a vertical axis labeled y.
Audio
Observation
The narrator introduces the independent variable as the x axis and the dependent variable as the y axis, saying he is making up the numbers.

Objects

Independent-variable table
Dependent-variable table
Horizontal x-axis
Vertical y-axis

Changes

Tables and axes are added one after another until the plotting framework is complete.

Invariants

x is consistently treated as the independent variable and y as the dependent variable.

Interpretation

The visual build establishes the two-variable structure needed before any regression line can be discussed.

Plotting the observations

Clear evidence

Shown in the video

Evidence

Diagram
Observation
Green dots appear one by one at (1,2), (2,4), (3,5), (4,4), and (5,5); the first two are temporarily annotated with coordinate labels.
Audio
Observation
The narrator says he will put in points for these observations and identifies the first as $x=1$ , $y=2$ and the second as point 2, 4.

Objects

Five green data points
Coordinate plane

Changes

The scatter plot fills in sequentially from left to right.

Invariants

Each plotted point corresponds to a paired row from the x and y tables.

Interpretation

The animation converts tabular paired data into a geometric scatter plot.

Mean lines and the required regression passing point

Clear evidence

Shown in the video

Evidence

Diagram
Observation
The word mean appears below the table with 3 under x and 4 under y. A vertical purple line is drawn at $x=3$ , a horizontal green line at $y=4$ , and their intersection is labeled (3,4). A dashed line passes through that point.
Audio
Observation
The narrator computes the means and says all regression lines have to go through that crossing point.

Objects

Vertical line $x=3$
Horizontal line $y=4$
Point (3,4)
Dashed sample regression line

Changes

Mean values are added below the table, then the corresponding mean lines are drawn, then the intersection point is labeled.

Invariants

The intersection remains ( $\bar{x}$ , $\bar{y}$ )=(3,4).

Interpretation

The graphic shows why the mean point is central: every regression line discussed must pass through it.

Visualizing deviations from the means

Clear evidence

Shown in the video

Evidence

Diagram
Observation
Dotted guide segments are drawn from each point horizontally toward the vertical mean line $x=3$ and vertically toward the horizontal mean line $y=4$ .
Audio
Observation
The narrator says he will take the distance from the x value to the mean for all observations and do the same thing for the y's.

Uncertainties

The spoken word "distance" is used while the later table records signed deviations rather than absolute distances.

Objects

Five data points
Mean lines $x=3$ and $y=4$
Dotted horizontal and vertical guide segments

Changes

Guide segments are added point by point to show each observation's offset from the mean lines.

Invariants

The mean lines stay fixed while the guides illustrate each observation's separation from them.

Interpretation

The animation motivates the deviation columns by showing horizontal offsets from x̄ and vertical offsets from ȳ.

Creating the x-deviation column

Clear evidence

Shown in the video

Evidence

Diagram
Observation
The graph simplifies, a new third column header x - x̄ is added to the table, and the entries -2, -1, 0, 1, 2 are filled in row by row.
Audio
Observation
The narrator says he will take the x value minus the mean at each point and then reads out each subtraction.

Objects

Expanded data table
Column header x - x̄
Numeric entries -2, -1, 0, 1, 2

Changes

The table expands from two columns to three, and each new cell is populated in sequence.

Invariants

The original x and y values remain unchanged while the deviation column is appended.

Interpretation

This step turns the geometric idea of offset from the mean line into explicit algebraic quantities for later regression formulas.

Drawing the horizontal y-mean line

Clear evidence

Shown in the video

Evidence

Diagram
Observation
A green horizontal line is drawn across the scatter plot at $y=4$ .
Audio
Observation
The speaker says he is drawing the mean line for the y values, which is mean of four.

Objects

Scatter plot axes
Five data points
Green horizontal line at $y=4$
Table with x, y, and $x - \bar{x}$ columns

Changes

A horizontal reference line appears at the y-mean value 4.

Invariants

The plotted data points do not move.
The x-axis and y-axis scales remain fixed.

Interpretation

The visual establishes $\bar{y}=4$ as the baseline from which vertical deviations $y-\bar{y}$ will be measured.

Visualizing y-deviations from the mean line

Clear evidence

Shown in the video

Evidence

Diagram
Observation
Dashed vertical segments connect individual points to the horizontal mean line while the y-deviation entries are filled.
Formula
Observation
The $y - \bar{y}$ column is completed row by row.

Uncertainties

The exact moment each dashed segment appears is approximate because the animation is continuous.

Objects

Data points
Horizontal mean line
Dashed vertical connectors
$y - \bar{y}$ column

Changes

Vertical dashed guides appear from points to the mean line.
The y-deviation column is filled with -2, 0, 1, 0, 1.

Invariants

The mean line stays at $y=4$ .
The original point coordinates stay unchanged.

Interpretation

The dashed segments show signed vertical distance from each observation to the mean, matching the numerical deviations in the table.

Introducing the regression line and target equation

Clear evidence

Shown in the video

Evidence

Diagram
Observation
A dashed straight line is drawn through the scatter plot.
Formula
Observation
The equation $\hat{y} = b_0 + b_1 x$ is written near the plot.

Uncertainties

The precise slope of the initially drawn dashed guide line is not numerically labeled on screen.

Objects

Scatter plot
Dashed straight line
Equation $\hat{y}=b_0+b_1x$

Changes

A candidate regression line is drawn through the points.
The symbolic model equation appears on screen.

Invariants

The data points remain the same five observations.
The axes and table layout remain visible.

Interpretation

The visual shifts from descriptive deviations to model fitting: the line represents the predicted relationship $\hat{y}=b_0+b_1x$ , and the next task is to determine $b_1$ .

Expanding the table for slope computation

Clear evidence

Shown in the video

Evidence

Formula
Observation
Two additional columns appear: ( $x - \bar{x}$ )^2 and ( $x - \bar{x}$ )( $y - \bar{y}$ ).
Diagram
Observation
The table expands to the right as values are entered row by row.

Objects

Original table columns
New squared-deviation column
New cross-product column

Changes

The squared x-deviation column is added and filled.
The cross-product column is added and filled.

Invariants

The original x, y, $x - \bar{x}$ , and $y - \bar{y}$ values remain unchanged.

Interpretation

The table is reorganized into the exact quantities needed for the least-squares slope: squared deviations for the denominator and cross-products for the numerator.

Summing columns and computing the slope

Clear evidence

Shown in the video

Evidence

Formula
Observation
The totals 10 and 6 are written beneath the corresponding columns.
Formula
Observation
The expression $b_1 = 6/10$ = .6 is displayed.

Objects

Squared-deviation column
Cross-product column
Column totals
Slope expression

Changes

The sum 10 is written under ( $x - \bar{x}$ )^2.
The sum 6 is written under ( $x - \bar{x}$ )( $y - \bar{y}$ ).
The slope expression resolves to 0.6.

Invariants

The underlying data points and earlier deviation values do not change.

Interpretation

The visual completes the algebraic transition from rowwise deviations to the single summary parameter $b_1$ , the slope of the regression line.

Visual construction of the slope formula

Clear evidence

Shown in the video

Evidence

Diagram
Observation
The table highlights the columns $(x-\bar{x})^2$ and $(x-\bar{x})(y-\bar{y})$ , with totals 10 and 6.
Animation
Observation
The formula under the table evolves from a numeric quotient to a symbolic summation quotient.

Objects

Scatter plot with five points
Dashed regression line
Data table with deviation columns
Formula $b_1=\frac{6}{10}=.6$

Changes

The numeric slope expression is rewritten first as a quotient of column expressions.
Then summation signs are added to produce $b_1=\frac{\sum(x-\bar{x})(y-\bar{y})}{\sum(x-\bar{x})^2}$ .

Invariants

The underlying data points remain the same.
The totals 10 and 6 remain fixed while the notation becomes more general.

Interpretation

The animation connects the concrete table arithmetic to the general least-squares slope formula.

Misconceptions · 5

Distance wording versus signed deviation entries

Approximate timing

Supplementary explanation

Evidence

Audio
Observation
The narrator says, "I'm going to take the distance from the x value to the mean" and later "take the distance each observation is from the y mean".
Diagram
Observation
The table column is labeled x - x̄ and contains negative values -2 and -1 as well as positive values 1 and 2.

Uncertainties

The video does not explicitly correct the wording; this item is an analyst clarification based on the visible signed entries.

Misconception

One may infer from the spoken phrase "distance" that the table should contain only nonnegative magnitudes.

Clarification

The displayed column $x-\bar{x}$ records signed deviations, because values such as $1-3=-2$ and $2-3=-1$ are negative.

Zero in either deviation makes the cross-product zero

Clear evidence

Derived from the video

Evidence

Formula
Observation
Rows with either $x-\bar{x}=0$ or $y-\bar{y}=0$ produce cross-product 0 in the table.
Audio
Observation
The speaker explicitly multiplies by zero several times while filling the product column.

Misconception

One might think every row contributes nonzero information to the slope numerator.

Clarification

In this method, if either $x_i-\bar{x}=0$ or $y_i-\bar{y}=0$ , then ( $x_i-\bar{x}$ )( $y_i-\bar{y}$ )=0, so that observation adds nothing to the numerator sum.

Squaring changes negative deviations into positive contributions

Clear evidence

Derived from the video

Evidence

Formula
Observation
Negative deviations -2 and -1 become 4 and 1 after squaring.
Audio
Observation
The speaker says to square all the values in the x-minus-mean column.

Misconception

One might expect negative x-deviations to reduce the denominator.

Clarification

Because the denominator uses ( $x_i-\bar{x}$ )^2, all terms are nonnegative; negative deviations still increase the sum after squaring.

Possible verbal imprecision about the denominator

Clear evidence

Shown in the video

Evidence

Audio
Observation
The speaker says he divides by “the difference between the x values and the mean squared,” while the displayed formula is $\sum(x-\bar{x})^2$ .
Formula
Observation
The on-screen symbolic form clarifies that the denominator is the sum of squared deviations, not the square of a single difference.

Misconception

A listener could infer from the wording that the denominator is one squared difference rather than a sum of squared deviations.

Clarification

The displayed formula resolves this: the denominator is $\sum(x-\bar{x})^2$ , the sum over all squared deviations of x from its mean.

Do not treat the preview as a full definition of R-squared

Clear evidence

Supplementary explanation

Evidence

Audio
Observation
The speaker only announces “Up next, I’m going to talk about R squared.”
Diagram
Observation
The preview labels distances but gives no formula.

Misconception

One might think this clip already defines $R^2$ mathematically.

Clarification

In this segment, $R^2$ is only introduced conceptually through distance labels; no complete formula or derivation is shown.

Concept relations · 14

Independent variable in the regression setup → Mean of the x-values

Clear evidence

Shown in the video

Evidence

Audio
Observation
The narrator first introduces x as the independent variable and then computes the mean of the x values.
Diagram
Observation
The x table precedes the appearance of mean 3 under the x column.

Proof dependency

Explanation

The mean $\bar{x}$ is computed directly from the listed independent-variable values.

Dependent variable in the regression setup → Mean of the y-values

Clear evidence

Shown in the video

Evidence

Audio
Observation
The narrator first introduces y as the dependent variable and then computes the mean of the y values.
Diagram
Observation
The y table precedes the appearance of mean 4 under the y column.

Proof dependency

Explanation

The mean $\bar{y}$ is computed directly from the listed dependent-variable values.

Mean of the x-values → Regression lines pass through the mean point

Clear evidence

Shown in the video

Evidence

Audio
Observation
The narrator says regression lines go through the point where the mean of the independent variable and the mean of the dependent variable cross.
Diagram
Observation
The intersection of $x=3$ and $y=4$ is labeled (3,4).

Application

Explanation

The computed value $\bar{x}=3$ supplies the x-coordinate of the mean point used in the claim.

Mean of the y-values → Regression lines pass through the mean point

Clear evidence

Shown in the video

Evidence

Audio
Observation
The narrator identifies the crossing point of the two means as the required point for regression lines.
Diagram
Observation
The horizontal mean line $y=4$ meets the vertical mean line $x=3$ at (3,4).

Application

Explanation

The computed value $\bar{y}=4$ supplies the y-coordinate of the mean point used in the claim.

Mean of the x-values → Deviation of each x-value from the mean

Clear evidence

Shown in the video

Evidence

Audio
Observation
The narrator says he will take the x value minus the mean at each point.
Diagram
Observation
The new column header is x - x̄ and each entry uses the previously established mean 3.

Proof dependency

Explanation

The deviation column cannot be formed without first knowing $\bar{x}$ .

Horizontal mean line for y → Deviations of y from its mean

Clear evidence

Shown in the video

Evidence

Audio
Observation
The speaker first identifies the mean of y as 4 and then subtracts 4 from each y value.
Formula
Observation
The $y - \bar{y}$ column depends on the stated mean $\bar{y}=4$ .

Prerequisite

Explanation

The horizontal mean line defines $\bar{y}=4$ , and that value is then used to compute each y-deviation.

Squared x-deviation column → Least-squares slope from deviation sums

Clear evidence

Shown in the video

Evidence

Formula
Observation
The slope expression uses the sums of ( $x - \bar{x}$ )^2 and ( $x - \bar{x}$ )( $y - \bar{y}$ ).
Audio
Observation
The speaker says the squared column gives the denominator and the product column gives the numerator.

Proof dependency

Explanation

The denominator of the slope formula is built from the squared x-deviation column.

Product of x- and y-deviations → Least-squares slope from deviation sums

Clear evidence

Shown in the video

Evidence

Formula
Observation
The numerator sum 6 comes from the ( $x - \bar{x}$ )( $y - \bar{y}$ ) column.
Audio
Observation
The speaker adds the product column and uses that total as the numerator.

Proof dependency

Explanation

The numerator of the slope formula is built from the cross-product column.

Linear regression target equation → Least-squares slope from deviation sums

Clear evidence

Shown in the video

Evidence

Formula
Observation
The model equation $\hat{y}=b_0+b_1x$ is introduced before $b_1$ is computed.
Audio
Observation
The speaker says the goal is to get that equation and then determines $b_1$ .

Contains

Explanation

The regression model contains the slope parameter $b_1$ , and the clip focuses on computing that parameter.

Least-squares slope formula → Linear regression equation form

Clear evidence

Shown in the video

Evidence

Formula
Observation
The slope formula is presented directly underneath the model $\hat{y}=b_0+b_1x$ .

Application

Explanation

The slope formula supplies the coefficient $b_1$ used in the regression equation.

Finding the intercept from the mean point → Least-squares slope formula

Clear evidence

Shown in the video

Evidence

Formula
Observation
The intercept calculation uses the already computed value $.6$ in $4=b_0+.6(3)$ .

Prerequisite

Explanation

Computing $b_0$ by substitution requires the previously found slope $b_1$ .

Regression line passes through the mean point → Finding the intercept from the mean point

Clear evidence

Shown in the video

Evidence

Audio
Observation
“The regression line has to cross point three four.”
Formula
Observation
That point is then substituted into the regression equation.

Proof dependency

Explanation

The method for finding $b_0$ depends on the claim that the fitted line passes through $(\bar{x},\bar{y})$ .

Find an answer · 16

What does x represent in this regression example?

Clear evidence

Shown in the video

Evidence

Audio
Observation
The narrator names the independent variable as the x axis.
Diagram
Observation
A table labeled independent variable lists x values.

Knowledge points

Independent variable in the regression setup

What does y represent in this regression example?

Clear evidence

Shown in the video

Evidence

Audio
Observation
The narrator names the dependent variable as the y axis.
Diagram
Observation
A table labeled dependent variable lists y values.

Knowledge points

Dependent variable in the regression setup

Why must the regression line pass through (3,4)?

Clear evidence

Shown in the video

Evidence

Audio
Observation
The narrator states that all regression lines have to go through the crossing point of the two means.
Diagram
Observation
The point (3,4) is marked at the intersection of the mean lines.

Knowledge points

Regression lines pass through the mean point
Mean of the x-values
Mean of the y-values

How is the x - x̄ column calculated for each observation?

Clear evidence

Shown in the video

Evidence

Audio
Observation
The narrator computes 1 minus 3, 2 minus 3, 3 minus 3, 4 minus 3, and 5 minus 3 aloud.
Diagram
Observation
The x - x̄ column is filled with -2, -1, 0, 1, 2.

Knowledge points

Deviation of each x-value from the mean
Computing the x-deviation column

Are the entries in x - x̄ distances or signed deviations?

Approximate timing

Supplementary explanation

Evidence

Audio
Observation
The narrator uses the word distance when describing offsets from the means.
Diagram
Observation
The table records signed values including negatives.

Uncertainties

This search target reflects an analyst-identified wording issue rather than an explicit correction in the video.

Knowledge points

Distance wording versus signed deviation entries
Deviation of each x-value from the mean

What is the mean of the y-values in this regression example?

Clear evidence

Shown in the video

Evidence

Audio
Observation
The speaker says the mean of y is 4 and draws the mean line.
Diagram
Observation
A horizontal line appears at $y=4$ .

Knowledge points

Horizontal mean line for y

How are the y-deviations $y - \bar{y}$ calculated from the table?

Clear evidence

Shown in the video

Evidence

Audio
Observation
The speaker computes 2 minus 4, 4 minus 4, 5 minus 4, etc.
Formula
Observation
The $y - \bar{y}$ column is filled with -2, 0, 1, 0, 1.

Knowledge points

Deviations of y from its mean

What regression equation is the video trying to determine?

Clear evidence

Shown in the video

Evidence

Formula
Observation
The equation $\hat{y} = b_0 + b_1 x$ is written on screen.
Audio
Observation
The speaker states the goal is to get this equation.

Knowledge points

Linear regression target equation

Why does the video square the $x - \bar{x}$ column?

Clear evidence

Shown in the video

Evidence

Audio
Observation
The speaker says to take the x-minus-x-bar column and square it.
Formula
Observation
The column ( $x - \bar{x}$ )^2 is added.

Knowledge points

Squared x-deviation column
Least-squares slope from deviation sums

What does the column ( $x - \bar{x}$ )( $y - \bar{y}$ ) represent?

Clear evidence

Shown in the video

Evidence

Audio
Observation
The speaker multiplies the x-deviation column by the y-deviation column.
Formula
Observation
The column ( $x - \bar{x}$ )( $y - \bar{y}$ ) is created and filled.

Knowledge points

Product of x- and y-deviations

How is the slope $b_1$ computed from the two column sums?

Clear evidence

Shown in the video

Evidence

Formula
Observation
The displayed calculation is $b_1 = 6/10$ = .6.
Audio
Observation
The speaker says the denominator is 10, the numerator is 6, and $b_1$ is 0.6.

Knowledge points

Least-squares slope from deviation sums
Computed slope for the example

What is the least-squares formula for the slope b1?

Clear evidence

Shown in the video

Evidence

Formula
Observation
Symbolic slope formula is shown explicitly.

Knowledge points

Least-squares slope formula
Derivation of the slope from the deviation table

Coverage and review notes

Covered · Title card and brief preview of the finished regression worksheet, with narration introducing regression.

Covered · Construction of the x and y tables and the coordinate axes.

Covered · Plotting the five paired observations as a scatter plot.

Covered · Computation of the means, drawing of mean lines, and statement that regression lines pass through (3,4).

Covered · Visual explanation of taking offsets from the mean lines; wording about distance is clarified by the later signed table entries.

Covered · Creation of the x - x̄ column and row-by-row numerical computation of its values.

Covered · The clip introduces the horizontal mean line for y at $\bar{y}=4$ .

Covered · The y-deviation column is computed row by row from the mean line.

Covered · Brief transition with no new mathematical statement beyond the completed deviation table.

Covered · The regression model $\hat{y}=b_0+b_1x$ is introduced and the task shifts to finding $b_1$ .

Covered · The squared x-deviation column is added and filled.

Covered · The cross-product column is added and filled row by row.

Covered · Column sums are taken and the slope $b_1=0.6$ is computed.

Covered · The clip opens with the table-based computation of the slope and its symbolic formula.

Covered · The speaker transitions from b1 to b0 and identifies b0 as the y-intercept.

Covered · The mean point (3,4) is marked and substituted into the regression equation.

Covered · The algebraic solution for b0 is written out step by step.

Covered · The coefficients are summarized and the final regression equation is displayed.

Covered · A title card and distance diagram introduce the next topic, R-squared, without giving its formula.

Covered · Closing branding/outro screen with no additional mathematical content.

Explore the knowledge in this video

Reviewed subject paths

Least squares fitting

Questions this video answers

Meet the concept

Find a method

Find a method

Meet the concept

Find a method

Understand why

Meet the concept

Understand why

How to calculate linear regression using least square method

The explanation, unpacked.

Chapters

Learning script

Knowledge cards

Independent and dependent variables in the example

Scatter plot of the five observations

Means of x and y

Regression lines pass through the mean point

From mean offsets to the deviation column

Computed x-deviations for the five rows

Horizontal mean line for y

Computing y-deviations

Target regression model

Squared x-deviation column

Cross-product deviation column

Slope formula from column sums

Least-squares slope formula

Regression line passes through the mean point

Solving for the intercept b0

Final fitted regression equation

R-squared preview only

Detailed learning notes

x

y

xˉ\bar{x}xˉ

yˉ\bar{y}yˉ​

x−xˉx-\bar{x}x−xˉ

x

y

xˉ\bar{x}xˉ

yˉ\bar{y}yˉ​

y^\hat{y}y^​

b0b_0b0​

b1b_1b1​

Independent variable in the regression setup

Dependent variable in the regression setup

Mean of the x-values

Mean of the y-values

Deviation of each x-value from the mean

Horizontal mean line for y

Deviations of y from its mean

Linear regression target equation

Squared x-deviation column

Product of x- and y-deviations

Least-squares slope from deviation sums

Linear regression equation form

Regression lines pass through the mean point

Computed slope for the example

Sum of squared x-deviations

Sum of cross-products

Regression line passes through the mean point

Standard textbook form of the slope

Computing the x-deviation column

Derivation of the y-deviation column

Derivation of squared x-deviations

Derivation of the cross-product column

Derivation of the slope b1b_1b1​

Derivation of the slope from the deviation table

Derivation of the intercept using the mean point

Assembling the final regression equation

Worked setup for a least-squares regression example

Worked least-squares slope example with five points

Worked least-squares regression example

Opening title and final-result preview

Construction of axes and data tables

Plotting the observations

Mean lines and the required regression passing point

Visualizing deviations from the means

Creating the x-deviation column

Drawing the horizontal y-mean line

Visualizing y-deviations from the mean line

Introducing the regression line and target equation

Expanding the table for slope computation

Summing columns and computing the slope

Visual construction of the slope formula

Distance wording versus signed deviation entries

Zero in either deviation makes the cross-product zero

Squaring changes negative deviations into positive contributions

Possible verbal imprecision about the denominator

$\bar{x}$

$\bar{y}$

$x-\bar{x}$

$\bar{x}$

$\bar{y}$

$\hat{y}$

$b_0$

$b_1$

Derivation of the slope $b_1$

How are the y-deviations $y - \bar{y}$ calculated from the table?

Why does the video square the $x - \bar{x}$ column?

What does the column ( $x - \bar{x}$ )( $y - \bar{y}$ ) represent?

How is the slope $b_1$ computed from the two column sums?