Scribd
Upload
16 views39 pages

Chapter 3 Describing Relationships

Uploaded by

王一荣
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
16 views39 pages

Chapter 3 Describing Relationships

Uploaded by

王一荣
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 39

Describing

Relationship
ILO(Intended Learning Outcomes)
This section covers…

• Scatter diagram and correlation

• Least-Squares Regression

• Correlation and regression


1. Explanatory and Response Variables

Response variable and explanatory variable

• A response variable measures an outcome of a study. An explanatory


variable may help explain or predict changes in a response variable.

① alcohol on body temperature

② driving speed on accident rate

③ age on the intelligence

• Explanatory variable explains the Response variable’s response


1. Explanatory and Response Variables
2. Displaying Relationships: Scatterplots

• A scatterplot shows the relationship between two quantitative variables


measured on the same individuals.

• We usually put the explanatory variable and the response variable

Step 1. Decide which variable should go on each axis.

Step 2. Label and scale your axes.

Step 3. Plot individual data values


2. Displaying Relationships: Scatterplots

Make a scatterplot of the relationship between


points per game and wins.
2. Displaying Relationships: Scatterplots

To describe a scatterplots

I. Direction (upper, lower, left, right, spread out)

II. Form (linear, curved, clusters)

III. Strength (strong, moderate, weak, hardly)

IV. Outliers
2. Displaying Relationships: Scatterplots

• the overall pattern moves from


lower left to upper right.

(This means that teams that score


more points per game have more
wins and teams that score fewer
points per game have fewer wins)
2. Displaying Relationships: Scatterplots

• There seems to be a linear


pattern in the graph (that is, the
overall pattern follows a straight
line).
2. Displaying Relationships: Scatterplots

• Because the points do not vary


much from the linear pattern, the
relationship is fairly strong.

• There do not appear to be any


values that depart from the linear
pattern, so there are no outliers.
2. Displaying Relationships: Scatterplots

• Therefore, the two variables have


strong positive association.
2. Displaying Relationships: Scatterplots

We need to use a numerical measure how strong a linear relationship is

• The correlation r measures the direction and strength of the linear


relationship between two quantitative variables.

• Correlation is a special association


2. Displaying Relationships: Scatterplots

• Examples of the strength’ of the correlation:


2. Displaying Relationships: Scatterplots

Calculating Correlation

This indicates

1. Correlation makes no distinction between explanatory and response


variables.

2. does not change when we change the units of measurement of x, y, or


both

3. The correlation itself has no unit of measurement.


2. Displaying Relationships: Scatterplots

The Property of correlation

1. Correlation requires that both variables be quantitative

2. Correlation are affected by a few outlying observations

3. Correlation does not imply causation. (Connectioncause and effect)

4. Correlation does not describe curved relationships between variables

5. A value of or does not guarantee a linear relationship between two variables.

6. Correlation is not a complete summary of two-variable


2. Displaying Relationships: Scatterplots
2. Displaying Relationships: Scatterplots
3. Regression line and Prediction

• A regression line is a line that describes how a response variable y


changes as an explanatory variable x changes.

We often use a regression line to


predict the value of y for a given
value of x.
3. Regression line and Prediction

Regression line, predicted value, slope, y intercept

• A regression line relating y to x has an equation of the form

where

is the predicted value of the response variable y

m is the slope,

c is the y intercept
3. Regression line and Prediction

Extrapolation

• Extrapolation is the use of a regression line for prediction far outside the
interval of values of the explanatory variable x used to obtain the line.
Such predictions are often not accurate.
3. Regression line and Prediction-Practice
4. Residuals and the Least-Squares Regression Line

Residual
• A residual is the difference between an observed value of the response
variable and the value predicted by the regression line. That Is,
4. Residuals and the Least-Squares Regression Line

Least-squares regression line

• The least-squares regression line of y on x is the line that makes the sum
of the squared residuals as small as possible
4. Residuals and the Least-Squares Regression Line
4. Residuals and the Least-Squares Regression Line

Residual plots
• A residual plot is a scatterplot of the residuals against the explanatory
variable. Residual plots help us assess whether a linear model is
appropriate.
4. Residuals and the Least-Squares Regression Line

• When an obvious curved pattern exists in a residual plot, the model we


are using is not appropriate
4. Residuals and the Least-Squares Regression Line

• Because there is only


random scatter in the
residual plot, we know the
linear model we used is
appropriate.
4. Residuals and the Least-Squares Regression Line

The Standard Deviation of the Residuals

• This value gives the approximate size of a “typical” prediction error (residual).
4. Residuals and the Least-Squares Regression Line
The Coefficient of Determination

This indicates the proportion of variation governed by the regression line.


If all the points fall directly on the least-squares line,
In the worst-case scenario,
4. Residuals and the Least-Squares Regression Line
4. Residuals and the Least-Squares Regression Line
4. Residuals and the Least-Squares Regression Line

• Interpreting Computer Regression Output


4. Residuals and the Least-Squares Regression Line
• A random sample of 15 high school students was selected from the U.S.
CensusAtSchool database. The foot length (in centimeters) and height (in
centimeters) of each student in the sample were recorded. Least-squares
regression was performed on the data. A scatterplot with the regression line added,
a residual plot, and some computer output from the regression are shown below.
4. Residuals and the Least-Squares Regression Line
4. Residuals and the Least-Squares Regression Line
4. Residuals and the Least-Squares Regression Line

Outliers and influential observations in regression

• An outlier is an observation that lies outside the overall pattern of the


other observations.

• An observation is influential for a statistical calculation if removing it


would markedly change the result of the calculation.

• Points that are outliers in the x direction of a scatterplot are often


influential for the least-squares regression line.
4. Residuals and the Least-Squares Regression Line

The best way to verify that a point is influential is to find the


regression line both with and without the unusual point.
If the line moves more than a small amount when the point is
deleted, the point is influential
4. Residuals and the Least-Squares Regression Line

Correlation and Regression Wisdom

• When you use these tools, you should be aware of their limitations:

• The distinction between explanatory and response variables is important


in regression but does not matter the correlation

• Correlation and regression lines describe only linear relationships.

• Correlation and least-squares regression lines are not resistant.

• Association does not imply causation.


4. Residuals and the Least-Squares Regression Line

• Technology

• Calculate the correlation

• Determine the equation of least-squres regression line

You might also like