Thanks I got it working. My first mistake was in the change I made to the R script. I exported the Gretl Data to a .csv file and just ran MINE this way rather then using Gretl GNU R. Now I can import the results back to Gretl for any additional analysis or plotting. source("MINE.R") MINE("c:\\gretldata.csv","all.pairs",0) MINE("c:\\gretldata.csv",0) #will both run MINE on "gretldata" #and have it analyze each variable only #against the 0-th variable. MINE("c:\\gretldata.csv","one.pair",0,5) MINE("c:\\gretldata.csv",0,5)# #will both run MINE on "gretldata" #and have it analyze only the 0-th #variable against the 5-th variable. ---------------------------------------------------------------------- Message: 1 Date: Tue, 3 Jan 2012 15:22:52 -0800 (PST) From: Ronald McEwan <ronmac17(a)yahoo.com&gt; Subject: [Gretl-users] Help with using MINE wirh Gretl To: &quot;gretl-users(a)lists.wfu.edu&quot; <gretl-users(a)lists.wfu.edu&gt; Message-ID:     <1325632972.23941.YahooMailNeo(a)web162102.mail.bf1.yahoo.com&gt; Content-Type: text/plain; charset="iso-8859-1" I have recently been working with "MINE" (http://www.exploredata.net/). MINE is a java program that will work with R. Now that Gretl works with R (and flawlessly I must say) I would like to use MINE with Gretl via R. So far I am able to run MINE as a command line function, but not in R. I am not able to make the necessary adjustment to the MINE.r code to get this to run. Hopefully someone with more experience with R can provide some advice in making the MINE.r code adjustments I need to make. This example is from the MINE.r example file that I tried to modify. I get "Error: could not find function "MINE". I think part of the problem is the change I made in line 2. When I run Gretl>tools>Start GNU R this puts me in c:\Gretl and I have MINE in c:\ . Do I have to put MINE in c:\Gretl or change something else? Thanks, Ron McEwan library("rJava") .jinit("c:\\MINE.jar") MINE("gretldata","all.pairs",0) MINE("gretldata",0) #?? will both run MINE on "gretldata" #?? and have it analyze each variable only #?? against the 0-th variable. # MINE("gretldata","one.pair",0,5) MINE("gretldata",0,5) Spellman#?? will both run MINE on "gretldata" #?? and have it analyze only the 0-th #?? variable against the 5-th variable. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.wfu.edu/pipermail/gretl-users/attachments/20120103/4c983b13/... ------------------------------ Message: 2 Date: Wed, 4 Jan 2012 12:38:21 +0000 (GMT) From: Allin Cottrell <cottrell(a)wfu.edu&gt; Subject: Re: [Gretl-users] Help with using MINE wirh Gretl To: Ronald McEwan <ronmac17(a)yahoo.com&gt;, Gretl list     <gretl-users(a)lists.wfu.edu&gt; Message-ID:     <alpine.DEB.2.00.1201041232150.32303(a)localhost6.localdomain6&gt; Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed On Tue, 3 Jan 2012, Ronald McEwan wrote: The R function "MINE" is defined in MINE.r, so you'll have to source() that file, or copy the required functions into an R file of your own. On a quick experiment, it appears that the R/java stuff doesn't work if gretl calls R via its shared library (on Linux at any rate). I found it necessary to do   set R_lib off in my gretl script before the "foreign" block that calls for the MINE analysis. Allin Cottrell ------------------------------ _______________________________________________ Gretl-users mailing list Gretl-users(a)lists.wfu.edu http://lists.wfu.edu/mailman/listinfo/gretl-users End of Gretl-users Digest, Vol 60, Issue 2 ******************************************