Exporting Multiple Linear Regression from Polymath to Excel

See Also: Exporting from Polymath to Excel  Variables and Expressions 

The "Export to Excel"  capability in the Data Table is executed by clicking on the Excel icon found on the particular regression window.  You must have your Excel program active on your desktop before exporting a program to Excel.  A problem should be exported after it has been completely entered successfully into Polymath and ready for solution. It is also good practice to also solve the problem in Polymath so that the solution in Excel can be compared and verified.  Upon export, the problem in Polymath will be completely transferred into Excel.  This transfer automatically includes the translation of logical statements and the intrinsic functions from Polymath into equivalent functions within Excel. 

Multiple Linear Regression

Example: Fitting a Multiple Linear Model to Heat of Hardening of Portland Cement versus Weight Percent of Components Data

Consider the data set shown below that is from Example 4 - Heat of hardening in the Polymath REG Regression Program. The data set may be obtained within the Polymath REG program by clicking on the Examples button and holding until Example 4 : Heat of hardening is highlighted.  This should bring the data into the Polymath Data Table.

A mouse click on the Regression tab at the right of the Data Table followed by another mouse click on the Multiple linear tab should bring up the window shown below.

Select Dependent Variable: hard_heat and the  Independent Variables: Wpc1,Wpc2,Wpc3, and dWpd4.  (Note that you must hold down on the left mouse key when clicking on each variable name.)  This problem is to be solved with the "Through origin" option checked.  The Excel program must be open prior to the export of the Polymath program.  A single mouse click on the Excel icon   results in the automatic export and solution of the problem in a new worksheet in Excel as shown below.

The regression calculations within Excel agree with the results obtained with Polymath for this same problem as shown below in Example 4 - "Heat of hardening" from the Polymath REG Regression Program.

Model: hard_heat = a1*Wpc1 + a2*Wpc2 + a3*Wpc3 + a4*Wpc4

General
Number of independent variables = 4
Regression not including a free parameter
Number of observations = 13

Statistics

Note that the intrinsic function LINEST is used in the Excel solution of this multiple linear regression problem.

Multiple Linear Regression Graph (created in Excel from generated results)

Multiple Linear Regression Residuals (created in Excel from generated results)