Response to why do we need a drop down to perl or R in SAS
in [SAS]
|
Prev: Inter- rater agreement for echocardiography measurements in SAS
Next: fitting mixture of Weibull distributions
From: Michael Raithel on 10 Feb 2010 17:19 Dear SAS-L-ers, Dale McLerran posted the following: > > I would like to see better support for R from SAS. R does get > packages for many statistical methods often long before those > methods are available in SAS. However, naming meta-analysis as > one of the benefits of R is misleading. Meta-analysis is easy > to do in SAS using the MIXED procedure. > > I have previously posted on the use of PROC MIXED to perform a > meta-analysis. I don't have the time right now to summarize > the methods or what is available for further study on SAS-L. > Interested persons can search the SAS-L archives for some of > those posts. > Dale, at WUSS 2009, I saw a great demo of the R Interface to SAS IML Studio 3.2, but I can't seem to find it. There is documentation on the SAS web site at: http://support.sas.com/software/products/imlstudio/index.html#overview So, maybe the support for R from SAS _IS_ getting better! Dale, best of luck in all your SAS endeavors! I hope that this suggestion proves helpful now, and in the future! Of course, all of these opinions and insights are my own, and do not reflect those of my organization or my associates. All SAS code and/or methodologies specified in this posting are for illustrative purposes only and no warranty is stated or implied as to their accuracy or applicability. People deciding to use information in this posting do so at their own risk. +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Michael A. Raithel "The man who wrote the book on performance" E-mail: MichaelRaithel(a)westat.com Author: Tuning SAS Applications in the MVS Environment Author: Tuning SAS Applications in the OS/390 and z/OS Environments, Second Edition http://www.sas.com/apps/pubscat/bookdetails.jsp?catid=1&pc=58172 Author: The Complete Guide to SAS Indexes http://www.sas.com/apps/pubscat/bookdetails.jsp?catid=1&pc=60409 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ A lot of people like snow. I find it to be an unnecessary freezing of water. - Carl Reiner +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
From: Dale McLerran on 11 Feb 2010 00:02 --- On Wed, 2/10/10, Michael Raithel <michaelraithel(a)WESTAT.COM> wrote: > From: Michael Raithel <michaelraithel(a)WESTAT.COM> > Subject: Re: Response to why do we need a drop down to perl or R in SAS > To: SAS-L(a)LISTSERV.UGA.EDU > Date: Wednesday, February 10, 2010, 2:19 PM > Dear SAS-L-ers, > > Dale McLerran posted the following: > > > > > I would like to see better support for R from > SAS. R does get > > packages for many statistical methods often long > before those > > methods are available in SAS. However, naming > meta-analysis as > > one of the benefits of R is misleading. > Meta-analysis is easy > > to do in SAS using the MIXED procedure. > > > > I have previously posted on the use of PROC MIXED to > perform a > > meta-analysis. I don't have the time right now > to summarize > > the methods or what is available for further study on > SAS-L. > > Interested persons can search the SAS-L archives for > some of > > those posts. > > > Dale, at WUSS 2009, I saw a great demo of the R Interface > to SAS IML Studio 3.2, but I can't seem to find it. > There is documentation on the SAS web site at: > > http://support.sas.com/software/products/imlstudio/index.html#overview > > So, maybe the support for R from SAS _IS_ getting better! > > Dale, best of luck in all your SAS endeavors! > > > I hope that this suggestion proves helpful now, and in the > future! > > Of course, all of these opinions and insights are my own, > and do not reflect those of my organization or my > associates. All SAS code and/or methodologies specified in > this posting are for illustrative purposes only and no > warranty is stated or implied as to their accuracy or > applicability. People deciding to use information in this > posting do so at their own risk. > > +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ > Michael A. Raithel > "The man who wrote the book on performance" > E-mail: MichaelRaithel(a)westat.com > > Author: Tuning SAS Applications in the MVS Environment > > Author: Tuning SAS Applications in the OS/390 and z/OS > Environments, Second Edition > http://www.sas.com/apps/pubscat/bookdetails.jsp?catid=1&pc=58172 > > Author: The Complete Guide to SAS Indexes > http://www.sas.com/apps/pubscat/bookdetails.jsp?catid=1&pc=60409 > > +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ > A lot of people like snow. I find it to be an unnecessary > freezing of water. - Carl Reiner > +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ > Michael, As a matter of fact, I am going to be demonstrating the R interface from IML studio to folks in my workplace about 3 weeks from now. This is a good start, but this interface does require a non-standard SAS interface - the IML Studio product. Now, there may be good reason to promote the IML Studio product itself. But I would like to see an interface to R from a standard SAS interface. Dale --------------------------------------- Dale McLerran Fred Hutchinson Cancer Research Center mailto: dmclerra(a)NO_SPAMfhcrc.org Ph: (206) 667-2926 Fax: (206) 667-5977 ---------------------------------------
From: NordlDJ on 11 Feb 2010 01:26 > -----Original Message----- > From: SAS(r) Discussion [mailto:SAS-L(a)LISTSERV.UGA.EDU] On Behalf Of > Dale McLerran > Sent: Wednesday, February 10, 2010 9:03 PM > To: SAS-L(a)LISTSERV.UGA.EDU > Subject: Re: Response to why do we need a drop down to perl or R in SAS > > --- On Wed, 2/10/10, Michael Raithel <michaelraithel(a)WESTAT.COM> wrote: > > > From: Michael Raithel <michaelraithel(a)WESTAT.COM> > > Subject: Re: Response to why do we need a drop down to perl or R in SAS > > To: SAS-L(a)LISTSERV.UGA.EDU > > Date: Wednesday, February 10, 2010, 2:19 PM > > Dear SAS-L-ers, > > > > Dale McLerran posted the following: > > > > > > > > I would like to see better support for R from > > SAS. R does get > > > packages for many statistical methods often long > > before those > > > methods are available in SAS. However, naming > > meta-analysis as > > > one of the benefits of R is misleading. > > Meta-analysis is easy > > > to do in SAS using the MIXED procedure. > > > > > > I have previously posted on the use of PROC MIXED to > > perform a > > > meta-analysis. I don't have the time right now > > to summarize > > > the methods or what is available for further study on > > SAS-L. > > > Interested persons can search the SAS-L archives for > > some of > > > those posts. > > > > > Dale, at WUSS 2009, I saw a great demo of the R Interface > > to SAS IML Studio 3.2, but I can't seem to find it. > > There is documentation on the SAS web site at: > > > > http://support.sas.com/software/products/imlstudio/index.html#overview > > > > So, maybe the support for R from SAS _IS_ getting better! > > > > Dale, best of luck in all your SAS endeavors! > > > > > > I hope that this suggestion proves helpful now, and in the > > future! > > > > Of course, all of these opinions and insights are my own, > > and do not reflect those of my organization or my > > associates. All SAS code and/or methodologies specified in > > this posting are for illustrative purposes only and no > > warranty is stated or implied as to their accuracy or > > applicability. People deciding to use information in this > > posting do so at their own risk. > > > > > +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ > > Michael A. Raithel > > "The man who wrote the book on performance" > > E-mail: MichaelRaithel(a)westat.com > > > > Author: Tuning SAS Applications in the MVS Environment > > > > Author: Tuning SAS Applications in the OS/390 and z/OS > > Environments, Second Edition > > http://www.sas.com/apps/pubscat/bookdetails.jsp?catid=1&pc=58172 > > > > Author: The Complete Guide to SAS Indexes > > http://www.sas.com/apps/pubscat/bookdetails.jsp?catid=1&pc=60409 > > > > > +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ > > A lot of people like snow. I find it to be an unnecessary > > freezing of water. - Carl Reiner > > > +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ > > > > Michael, > > As a matter of fact, I am going to be demonstrating the R > interface from IML studio to folks in my workplace about 3 > weeks from now. This is a good start, but this interface > does require a non-standard SAS interface - the IML Studio > product. > > Now, there may be good reason to promote the IML Studio > product itself. But I would like to see an interface to R > from a standard SAS interface. > > Dale > > --------------------------------------- > Dale McLerran > Fred Hutchinson Cancer Research Center > mailto: dmclerra(a)NO_SPAMfhcrc.org > Ph: (206) 667-2926 > Fax: (206) 667-5977 > --------------------------------------- Dale, I don't have a solution to the lack of a standard interface from SAS to R, but there may be an alternative. I just started playing with the SAS ODBC driver that comes with SAS v9. After setting up the SAS ODBC driver and a local ODBC server I am able to read SAS datasets from R using the RODBC package. I haven't successfully written an R dataframe to a SAS dataset yet, but it is supposed to be possible. This will allow fairly easy use of R statistical methods with data stored in SAS datasets. All one needs is BASE SAS. Dan Daniel J. Nordlund Washington State Department of Social and Health Services Planning, Performance, and Accountability Research and Data Analysis Division Olympia, WA 98504-5204
From: xlr82sas on 11 Feb 2010 04:00
On Feb 10, 10:26 pm, Nord...(a)DSHS.WA.GOV ("Nordlund, Dan (DSHS/RDA)") wrote: > > -----Original Message----- > > From: SAS(r) Discussion [mailto:SA...(a)LISTSERV.UGA.EDU] On Behalf Of > > Dale McLerran > > Sent: Wednesday, February 10, 2010 9:03 PM > > To: SA...(a)LISTSERV.UGA.EDU > > Subject: Re: Response to why do we need a drop down to perl or R in SAS > > > --- On Wed, 2/10/10, Michael Raithel <michaelrait...(a)WESTAT.COM> wrote: > > > > From: Michael Raithel <michaelrait...(a)WESTAT.COM> > > > Subject: Re: Response to why do we need a drop down to perl or R in SAS > > > To: SA...(a)LISTSERV.UGA.EDU > > > Date: Wednesday, February 10, 2010, 2:19 PM > > > Dear SAS-L-ers, > > > > Dale McLerran posted the following: > > > > > I would like to see better support for R from > > > SAS. R does get > > > > packages for many statistical methods often long > > > before those > > > > methods are available in SAS. However, naming > > > meta-analysis as > > > > one of the benefits of R is misleading. > > > Meta-analysis is easy > > > > to do in SAS using the MIXED procedure. > > > > > I have previously posted on the use of PROC MIXED to > > > perform a > > > > meta-analysis. I don't have the time right now > > > to summarize > > > > the methods or what is available for further study on > > > SAS-L. > > > > Interested persons can search the SAS-L archives for > > > some of > > > > those posts. > > > > Dale, at WUSS 2009, I saw a great demo of the R Interface > > > to SAS IML Studio 3.2, but I can't seem to find it. > > > There is documentation on the SAS web site at: > > > >http://support.sas.com/software/products/imlstudio/index.html#overview > > > > So, maybe the support for R from SAS _IS_ getting better! > > > > Dale, best of luck in all your SAS endeavors! > > > > I hope that this suggestion proves helpful now, and in the > > > future! > > > > Of course, all of these opinions and insights are my own, > > > and do not reflect those of my organization or my > > > associates. All SAS code and/or methodologies specified in > > > this posting are for illustrative purposes only and no > > > warranty is stated or implied as to their accuracy or > > > applicability. People deciding to use information in this > > > posting do so at their own risk. > > > +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ > > > Michael A. Raithel > > > "The man who wrote the book on performance" > > > E-mail: MichaelRait...(a)westat.com > > > > Author: Tuning SAS Applications in the MVS Environment > > > > Author: Tuning SAS Applications in the OS/390 and z/OS > > > Environments, Second Edition > > >http://www.sas.com/apps/pubscat/bookdetails.jsp?catid=1&pc=58172 > > > > Author: The Complete Guide to SAS Indexes > > >http://www.sas.com/apps/pubscat/bookdetails.jsp?catid=1&pc=60409 > > > +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ > > > A lot of people like snow. I find it to be an unnecessary > > > freezing of water. - Carl Reiner > > > +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ > > > Michael, > > > As a matter of fact, I am going to be demonstrating the R > > interface from IML studio to folks in my workplace about 3 > > weeks from now. This is a good start, but this interface > > does require a non-standard SAS interface - the IML Studio > > product. > > > Now, there may be good reason to promote the IML Studio > > product itself. But I would like to see an interface to R > > from a standard SAS interface. > > > Dale > > > --------------------------------------- > > Dale McLerran > > Fred Hutchinson Cancer Research Center > > mailto: dmclerra(a)NO_SPAMfhcrc.org > > Ph: (206) 667-2926 > > Fax: (206) 667-5977 > > --------------------------------------- > > Dale, > > I don't have a solution to the lack of a standard interface from SAS to R, but there may be an alternative. I just started playing with the SAS ODBC driver that comes with SAS v9. After setting up the SAS ODBC driver and a local ODBC server I am able to read SAS datasets from R using the RODBC package. I haven't successfully written an R dataframe to a SAS dataset yet, but it is supposed to be possible. This will allow fairly easy use of R statistical methods with data stored in SAS datasets. All one needs is BASE SAS. > > Dan > > Daniel J. Nordlund > Washington State Department of Social and Health Services > Planning, Performance, and Accountability > Research and Data Analysis Division > Olympia, WA 98504-5204- Hide quoted text - > > - Show quoted text - Hi, R can read and write SAS V5 xport datasets. It is loseless on IBM mainframes. There ia a very slight loss of precision when converting IEEE(Unix and Windows) floating point to V5 xport format(IBM floating point). This is due to an 11 bit exponent and 53 bit mantissa in IEEE and and 8 bit expontent and 56 bit mantissa in IBM floating point. If only SAS allowed calls to Java fron FCMP we could call R functions from the datastep. SAS->Java->R->Java->SAS. Java would be fast because all it is doing is passing pointers. Dale, I think you are right about proc mixed for continuous outcomes, but for odds ratios with rare adverse events in 2x2 contingency tables the meta packages in R seems very robust. The real issue with rare AEs(ie events like Lou Gerhigs desease) is 0 and double 0 cells. R provides several continuity corrections ie cross-arm, proportional weights and . 5(and others). In addition arcsine transforms, Peto and Mantel Haenzel methods come to mind. Some highly respected scientists, like Dr Olkind(Stanford I think) were involved with one of the meta packages. I believe the main meta package in R was federally founded. A big issue I have with R is the lack of support when you need to know the details. There s a lot of talk about switching from mixed to glimmix where I work. It seems GLM->Mixed->GLIMMIX is a direction, I think it has to do with non-normal residuals and better handling of random effects. Regards |