Least-squares idea The table below gives a small set of...

Least-squares idea The table below gives a small set of data. Which of the following two lines fits the data better: $\hat{y}=1-x$ or $\hat{y}=3-2 x ?$ Make a graph of the data and use it to help justify your answer. (Note: Neither of these two lines is the least-squares regression line for these data.) $$ \begin{array}{lrllrr} \hline x: & -1 & 1 & 1 & 3 & 5 \\ y: & 2 & 0 & 1 & -1 & -5 \\ \hline \end{array} $$

Least-squares idea The table below gives a small set of data. Which of the following two lines fits the data better: $\hat{y}=1-x$ or $\hat{y}=3-2 x ?$ Make a graph of the data and use it to help justify your answer. (Note: Neither of these two lines is the least-squares regression line for these data.)
$$
\begin{array}{lrllrr}
\hline x: & -1 & 1 & 1 & 3 & 5 \\
y: & 2 & 0 & 1 & -1 & -5 \\
\hline
\end{array}
$$

Key Concepts

Model Comparison

When multiple models or lines are proposed, comparing them involves analyzing measures of fit, such as the sum of squared residuals, and using visual and numerical methods to judge which model more accurately captures the relationship between variables. This comparative approach is vital in selecting the most appropriate model.

Least-Squares Regression

This concept centers on determining the line that minimizes the sum of the squared differences between observed and predicted values. It provides a systematic method for finding the best-fitting line through a set of data points, even though in some cases, alternative lines might be compared qualitatively without strictly following the least-squares criterion.

Residual Analysis

Residuals, defined as the differences between observed values and the values predicted by a model, are key for assessing the accuracy of a fitted line. By examining these differences, one can determine which model better represents the data by identifying smaller and more uniformly distributed residuals.

Graphical Data Analysis

Graphing data points along with the candidate fitted lines is an important technique in exploratory data analysis. Visual inspection helps in understanding the overall trend, detecting potential outliers, and evaluating how closely each line follows the pattern of the data.

Transcript

00:01 We're given this table with a small set of data and which of the following two fits the lines better? does this or this fit better? we're gonna make a graph and use it.

00:28 Okay, so let's do a graph.

00:32 I think i'll go to the next page so i can use a big graph.

00:40 Fyi, it's hard to graph on this so so a little bit more room here.

00:59 Okay, and then i'm going to go.

01:06 Okay, so here's zero.

01:08 And then i'm going to go up by two, three, one, two, three.

01:19 Here i'm going to go half, two.

01:30 This is not going to be enough.

01:31 Let me try again.

01:35 Let's go by two.

01:38 One, two, three, four, five.

01:47 Okay, that's good enough here.

01:49 And here we have so this is 1 2 3 4 5 negative 1 negative 2 negative 3 and here i have 1 2 3 4 1 2 3 4 1 and 2 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 negative 1 negative 2 negative 3 and negative 4 and if i need to go farther i can um actually i do need to go further so let's go ahead and take these off let's do 3 1 2 3 4 5 and 1 2 1 one, two, three.

03:08 Okay, so let's begin here.

03:14 X, one minus y.

03:17 One minus, y equals one minus x.

03:21 Y equals one minus x.

03:30 So let's figure out if y equals one minus x.

03:38 If x equals zero, then y equals one.

03:45 And if, i don't know, what should we pick here? if y equals, shall we try, i don't think i'll hit a y zero there.

04:00 Let's pick a y equals, how about a negative, if y equals one, then x equals zero.

04:12 Okay, so i'm going to have for my y, let's pick two nice bright spots here.

04:18 This and one and zero and one and zero...