Skip to content

R integration functions

The R expressions let you enhance expression calculations using R functionality. You can access R functionality through two expression functions: RCall and RScript. You can find these functions in the Modelling Functions folder in the expression editor.

Note

The ModelScore expression function is also used in relation to R models — see Scoring records using an R model.

Note

To use the R expressions, you must first set up the R integration for FastStats.

RCall

This function allows you to call out to an R function with data from your FastStats system.

Expression
RCall(functionName, number1, number2, ...)

This calls a function named functionName in R with a vector of numbers c(number1, number2, ...).

Parameters

The first parameter (text) is the name of the function you want to call.

The remaining parameters (numeric) are numbers that you want to pass as arguments to the function call.

Examples

Access functionality not implemented in FastStats expressions — calculate the 0.975 quantile of the standard normal distribution (=1.96):

Expression
RCall("qnorm", 0.975)

Pass a vector of values to the R function — can include variables and calculated values:

Expression
RCall("mean", 10, [Number of Bookings], 200 + 20)

Note

If the function is not loaded automatically at the start of an R session and needs a library() call to make it available (for example, if it's from a third-party CRAN package), you will need to use RScript and include the library() call as an additional command before the function call itself. See the R integration for FastStats documentation for information on changing the default packages available on start-up.

RScript

This function allows you to call out to a number of R functions with data from your FastStats system, using an R script.

Expression
RScript(script, number1, number2, ...)

This runs script in R. @1, @2, etc. are substituted by number1, number2, etc. when the script is run.

Parameters

The first parameter (text) is the script you wish to run.

The remaining parameters (numeric) are numbers you wish to substitute into the placeholder values @1, @2, etc. in the script.

Examples

Call a single function using RScript to specify optional parameters — calculate the 0.9 quantile of a normal distribution with mean 80 and standard deviation 5 (=86.4078):

Expression
RScript("qnorm(@1,mean=@2,sd=@3)", 0.9, 80, 5)

Use multiple R commands:

Expression
RScript("x<-mean(@1,@2);y<-median(@1,@2,@3,@4);min(x,y)", [Pie sales], [Cake sales], [Nut sales], [Cheese sales])`

Note

As with normal text expressions, the script parameter for RScript is limited to 32767 characters in length.

Note

There are some security-related restrictions on what the script can do. For example, you cannot run an executable or access the computer's filesystem.