Is it possible to to pass in formulaes like z~x+y for a regression model ? about rprovider HOT 4 CLOSED

You can just pass the formula as a string.

from rprovider.

Well, I tried it, but I get an error, it complains that the objects are not found.

Example : (I just str the two objects to proff that thay are well created)

R.str(msft);;
num [1:254] -0.0041 0.01427 -0.02562 0.0134 -0.00548 ...
val it : SymbolicExpression =
RDotNet.SymbolicExpression {Engine = RDotNet.REngine;
IsClosed = false;
IsInvalid = false;
IsProtected = true;
Type = Null;}
R.str(goog);;
num [1:254] -0.00404 0.00383 -0.0208 -0.006 -0.00712 ...
val it : SymbolicExpression =
RDotNet.SymbolicExpression {Engine = RDotNet.REngine;
IsClosed = false;
IsInvalid = false;
IsProtected = true;
Type = Null;}
R.rq("msft~goog",0.5);;
Error in eval(expr, envir, enclos) : object 'msft' not found
RDotNet.ParseException: Error in the application.
at RDotNet.REngine.Parse(String statement, StringBuilder incompleteStatement) in C:\dev\Sierra\3rdParty\R.NET\src\R.NET\REngine.cs:line 455
at RDotNet.REngine.d__0.MoveNext() in C:\dev\Sierra\3rdParty\R.NET\src\R.NET\REngine.cs:line 365
at System.Linq.Enumerable.LastOrDefault[TSource](IEnumerable1 source) at RDotNet.REngine.Evaluate(String statement) in C:\dev\Sierra\3rdParty\R.NET\src\R.NET\REngine.cs:line 324 at RProvider.RInteropInternal.eval(String expr) in C:_Download\Dropbox\42.LocalCode_F#\RType Provider\FSharpRProvider\RInterop.fs:line 282 at RProvider.RInterop.callFunc(String packageName, String funcName, IEnumerable1 argsByName, Object[] varArgs) in C:_Download\Dropbox\42.LocalCode_F#\RType Provider\FSharpRProvider\RInterop.fs:line 383
at RProvider.RInterop.call(String packageName, String funcName, Object[] namedArgs, Object[] varArgs) in C:_Download\Dropbox\42.LocalCode_F#\RType Provider\FSharpRProvider\RInterop.fs:line 398
at <StartupCode$FSI_0011>.$FSI_0011.main@()
Stopped due to error

The very same line works well in R :

rq("googmsft",0.5)
Call:
rq(formula = "googmsft", tau = 0.5)

Coefficients:
(Intercept) msft
0.0007102887 0.3720877846

Degrees of freedom: 362 total; 360 residual

from rprovider.

OK, found it, I need to instantiate them on R side

R.assign("msft",msft)
R.assign("goog",goog)
let res = R.rq("msft~goog",0.5)

Thx a lot for this awesome library !

from rprovider.

Yes - if you don't pass some other object then R expects the names to be bound in the environment.

Another way of doing this is to pass in a data.frame as the data parameter of rq. It will then bind the names in the formula to the columns of the data frame.

from rprovider.