R Tutorials | Statistics and Data Analysis with R | #rstats

What is R?

Using R or R-Studio for Data Analysis and Statistics

#rstats


Installing R 

Linux, Windows, Mac

Rconsole, RStudio R online

Packages and Libraries

Basic commands in R

First and foremost, we need to be able to manipulate data in R.

1. Our first action would be to find a working directory or set a new one in R (functions getwd() and setwd() in R). The procedure is the following:

2. Next, we enter our data. In R, we can 1) enter the data manually or 2) load an already existing data file:

1) creating the dataset (function data.frame()) and entering the data manually (function edit()) in R:

2) loading a data file in R (reading a .csv or a .txt file) by means of the R function read.table():

We can enter either the entire address or just the name of the file provided that we have specified the working directory (see above - getwd() and setwd())

3. Having loaded the data file, we need to upload it (or 'attach') to the working memory of R - attach():

4. To open or visualise a file in R, we type its name in R:

5. We can analyse the structure of the data file by means of the R function str():

6. Finally, we can save the data file (e.g. in .csv format) by using the function write.csv():

7. Other important commands in R:

• to add a variable to a dataset, we can use the command cbind(): 

          cbind(mydata, weight)

• to recode a variable, we use the function recode(): 

          country<- recode(region, '"OECD"=1; else=2; ', as.factor.result=TRUE)

• to order information in a file in an increasing/decreasing order with respect to a variable, we use the command order():

          mydata[order(-weight),]

• to create a subset of a dataset, we use the command subset():

          OECD<-subset(mydata, subset=(region=="OECD"))

• to define a variable as a label variable, we use the function row.names():

          row.names(mydata)<-number

Using R for Statistics

1. Descriptive Statistics with R

2. Linear and Multivariate Regression with R

3. Principal Component Analysis (PCA) with R

4. Hierarchical Clustering Models (HCM) with R

More info about R as well as other R language tutorials you can find on the homepage of the R programming language - r-project.org as well as on r-bloggers.com, a website which accumulates the knowledgebase of the R programming language all over the net. You are also very welcome to subscribe to the R-bloggers' podcast - http://feeds.feedburner.com/RBloggers.