BC Statistics 12 Linear Models Activity: Linear model fitting
(0)An activity exploring properties of a linear model fitted to bivariate data and introducing linear model fitting via software.
Prerequisite Knowledge
- Define the concept of least squares estimation in linear models.
- Interpret a sample correlation.
Learning Objectives
- Identify which variable is most naturally the response variable in a regression analysis.
- Fit a linear model to a bivariate data set via software.
- Recall that the centroid of the bivariate data, as defined by the coordinate given by the sample means, lies on the least squares line.
Description
The activity defines the regression line for bivariate data and explores aspects of a least squares line fit for fitting a linear model. An applet is used that fits a linear model to a randomly generated scatterplot. Linear model fitting is applied via several software applications such as R, Microsoft Excel, and an online applet, to carry out the model fitting and generate graphical presentations. The activity is designed to use with high school students but can be easily modified to use with students at other levels.
Creator
- Dunham, Bruce
Resource Type
Date Created
Date Approved
Access
Everyone
This resource is licensed under
Creative Commons Attribution 4.0 International (CC BY 4.0)
Please use this identifier to cite or link to this resource:
https://statspace.elearning.ubc.ca/handle/123456789/631
Comments
Ratings in detail