Skip to content

Database

5 messages · Trying To learn again, R. Michael Weylandt, Brian Ripley +2 more

#
I don't use Access but my general impression is that the advantages it
brings will be similar to those brought by any other database:
performance rather than ability -- they are both Turing complete after
all, after some trickery on the SQL end.

Databases allow much larger data sets than R currently does and often
allow faster queries -- some would argue the SQL syntax is clearer for
some subsetting operations, but that's perhaps a function of
familiarity. For the task you describe, it should be elementary in
both platforms and I'd just use whichever one the data was already in.
For more substantive data analysis, you almost certainly want to use
R.

Others with Access experience (or more SQL) can add more.

Michael

On Tue, Feb 28, 2012 at 6:06 PM, Trying To learn again
<tryingtolearnagain at gmail.com> wrote:
#
On 29/02/2012 12:45, R. Michael Weylandt wrote:
But not much larger in Access than 64-bit R allows: Access is a pretty 
limited system.
Access does not bring the performance benefits of more advanced DBMS 
engines: on Windows I would certainly recommend using SQL Server Express 
(or whatever it is currently called) instead.  If you want to use a DBMS 
to supplement R (as per the R Data Import/Export manual) I would use 
MySQL or SQLite.
'files'?  Where are the data?  If you mean different tables in one 
Access database then I would still do this in R via RODBC.
In which case you can simply use RODBC to import tables to R and work there.

  
    
#
On Tue, Feb 28, 2012 at 6:06 PM, Trying To learn again
<tryingtolearnagain at gmail.com> wrote:
The examples section at the bottom of ?sqldf in the sqldf package
shows how to do a variety of calculations in both SQL and in R.  Also
see the home page at http://sqldf.googlecode.com
#
It all depends on what you are doing but R is pretty powerful.  I have never used Access so I don't know what it can do but I have played around with othe dbs at a very basic level and most things I did could be done quite easily in R  : Sheer data set size could be a problem but unless you have millions of data items you are probably okay in R.

For your example.  Assume nams1 is from dataset 1 and nams 2 is from dataset 2.
=================================
nams1 <- letters[1:5]
nams2 <- letters[3:7]

nams1 %in% nams2
=====================================
Done.

John Kane
Kingston ON Canada
____________________________________________________________
Share photos & screenshots in seconds...
TRY FREE IM TOOLPACK at http://www.imtoolpack.com/default.aspx?rc=if1
Works in all emails, instant messengers, blogs, forums and social networks.