Loading

Module 1: Residual analysis

Notes
Study Reminders
Support
Text Version

Plotting the residuals

Set your study reminders

We will email you at these times to remind you to study.
  • Monday

    -

    7am

    +

    Tuesday

    -

    7am

    +

    Wednesday

    -

    7am

    +

    Thursday

    -

    7am

    +

    Friday

    -

    7am

    +

    Saturday

    -

    7am

    +

    Sunday

    -

    7am

    +

XSIQ
*

Core Mathematics - Plotting the residuals

Plotting the residuals

Plotting residuals

The residuals can then be plotted on a scatterplot to determine whether or
not a linear relationship adequately describes the data.

CASE 1 (residual plot appears to be random)

* No pattern exists. The residuals are randomly spread around the zero
line.

* There is no reason to look beyond a linear model.

CASE 2 (pattern exists)

* A pattern exists therefore there may be a better model than a _linear
model_.

* A suitable transformation may be able to linearise the data.

From previous example y = 1.17x + 0.42

* No obvious pattern exists, therefore the linear model (y = 1.17x +
0.42) is accepted.

Previous | Next