Predictions using smooth.spline() in TERR can differ from predictions using smooth.spline() in open-source R.  The TERR implementation of smooth.spline() is not the same as the open-source R implementation.

This is documented in the help file that is opened by the following command in a TERR Console session:

?smooth.spline

The following section of the help file for smooth.spline() documents the difference:

=============
Differences Between TIBCO Enterprise Runtime For R And Open-Source R

This function is based on the S-PLUS implementation and has not been completely converted to the R parameterization. The spar parameters do not match: They are on the same scale, but differ by a constant. Several arguments used in the R implementation are included here but not used. This implementation might choose a slightly rougher spline fit than the implementation in open-source R.
=============

If you need to use the open-source R version of smooth.spline() in a TERR script, you can use the REvaluate() function from TERR's optional built-in "RinR" package to send data and commands out to an open-source R instance that you have installed on the same machine.

The following public knowledge base articles provide information about using the RinR::REvaluate() function:

https://support.tibco.com/s/article/Data-function-s-TERR-script-to-send-commands-to-open-source-R-fails-with-error-could-not-find-function-REvaluate

https://support.tibco.com/s/article/How-to-use-the-data-and-expr-arguments-in-the-REvaluate-function-from-TERR-s-RinR-package-to-send-data-and-expressions-out-to-an-open-source-R-session

https://support.tibco.com/s/article/TERR-TSSS-or-a-Spotfire-data-function-returns-Error-in-REvaluate-there-is-no-package-called