Is there a table comparing SQL commands with R commands?

Question

Or a list of how to do in R, what are you doing in SQL (or vice versa)?

Thanks,

Tal

Answer 1

sqldf package maybe maybe some help here?

There is also a conversation from Joshua's accompanying presentation that Shane mentioned.

Answer 2

You can watch Joshua Reich's presentation on R and SQL (see page 11).

Answer 3

In the examples section at the bottom of the help page (sqldf) in the sqldf package, there are quite a few SQL commands and their mappings R.

Answer 4

Also worth a look at the RMysQL package.

I work with very large datasets that cannot be dumped into text before importing into R. This package allows me to use standard mysql queries from R to pull subsets of my data.

Answer 5

I recently started working with RMySQL and really liked the package. I just run basic SQL queries in R. Most of the data reinstallation is done in several independent SQL scripts, mainly some stored procedures.

I think R is a statistical package with some good merge capability, but it is not designed to handle relational data in this way. I work a lot with microdata and have to set up non-relational datasets from these microdata (and then use R to do regression analysis and build ggplot2 (!)). I also perform data aggregation in SQL just before connecting to R.

I also recommend using views (if they are fast enough for you). R refers to them as regular tables, using the list table operator.

There is also RPostgreSQL if you want to try postgreSQL. I tried this once, but switched to RMySQL because RPostgreSQL was so hard to configure on my Mac, and after updating the config disappeared. RMySQL was a lot easier. Then I had to compile the package myself, so if you run another OS, you can get a binary file (or there is a Mac OS).

In any case, there is some RPostgreSQL literature that can help you even if you use RMySQL, especially if you plan to use it for timers data (for example, TSPostgreSQL).