Multiple Linear Regression

macOS / Education

This app will compute a Multiple Linear Regression using up to six independent variables (x1 to x6). The general equation form for a multiple linear regression is:

y = b(0) + b(1)x1 + b(2)x2 + b(3)x3 + b(4)x4 + b(5)x5 + b(6)x6

where b(0) to b(6) are coefficients.

y is the dependent variable

x1 to x6 are the independent variables and

b() are the computed coefficients

This app allows for an infinite number of data points for each regression. Example datasets are included to help users get started quickly. These examples serve as guides for organizing and entering data. You can turn these examples off under Preferences. A main menu provides the organizational structure for this app. From this menu you can choose to start a new project, work on an existing project (such as data entry), or produce tables and graphs for a project. Menu items can also be reached from buttons on the toolbar.

Each new regression is organized as a project. This organizational structure follows throughout all of the regression and ANOVA apps in the Computational Suite. New projects are created by assigning a unique ID to the project. After entering the ID you can enter details associated with the project along with data. Data can be entered manually or from a comma delimited text file.

Reports include the regression variables, data frequency and residuals. If you have made any data entry errors, simply go back and correct them, then click on the reports menu item again.

Both 2D and 3D graphics are available. You can test the sensitivity of the beta coefficients with the up/down arrows. For the 3D graph, use the Zoom, Yaw, Pitch and Roll controls to see the entire graph from different angles.

This app is also equipped with an automatic backup system. Each time you quit the app a backup copy is generated. Up to 1000 backup copies are kept. To restore a project from a prior backup, select the last menu item and follow the instructions.

Quoi de neuf dans la dernière version ?

Multiple Linear Regression along with other apps in the Regression Suite have been re-written to be more flexible and user friendly. The general appearance and functionality of all regression apps (Linear Regression, Multiple Linear Regression, CurviLinear Regression, Multiple CurviLinear Regression) has been made uniform between apps and more intuitive for the user.