By writing the Great Wall poem October 30, 2011 Comments Off
First, data import
For beginners, facing a blank command-line window, the first real difficulty may be the import of data. There are many ways to import data, such as fetching from Web pages, obtaining public data sources, and importing text files. For a quick start, it is recommended that beginners take the R language in collaboration with Excel spreadsheet methods. That is, you first read and organize the data you want to work with the more familiar Excel, and then "paste" into R.
For example, we first download the Iris.csv demo data from this address, open it in Excel, select all the samples and then "copy". Enter the following command in the R language:
data=read.table (' clipboard ', T)
The read.table is a common command for R to read external data, t means that the first row is the header information, and the entire data exists in a variable named data. Another convenient way to import is to take advantage of the Rstudio function, and choose "Import DataSet" from the workspace menu.
Second, dataframe operation
After the data is imported into the R language, it is stored as a data frame (dataframe). Dataframe is an R data format that can be thought of as a statistical table, where each row represents a sample point, and each column represents a different property or feature of the sample. The basic operation method that beginners need to Master is Dataframe's editing, extracting and arithmetic.
Although it is recommended that beginners work with the data in Excel, sometimes it is necessary to edit the data in R, the following command gives you the opportunity to modify the data and deposit it into the new variable NewData:
newdata=edit (data)
Another scenario is that we may only focus on part of the data, such as extracting the sepal.width variable data from the sample number 20th to 30th from the original data, because the Sepal.width variable is the 2nd variable, so type the following command at this point:
newdata=data[20:30,2]
If you need to extract the sepal.width variable for all data, the following two commands are equivalent:
newdata=data[,2] Newdata=data$sepal.width
The third case is the need to perform some operations on the data, such as the need to enlarge the sepal.width variables of all the samples 10 times times, we first copy the original data, and then use the $ symbol to extract the operand:
Newdata=data newdata$sepal.width=newdata$sepal.width*10
Iii. Description of statistics
Descriptive statistics is a tool to extract information from a large amount of data, the most common is the summary command, run Summary (data) results are as follows: five sub-sites and mean values are computed for numeric variables, and frequency is calculated for categorical variables.
The mean and standard deviation of the sepal.width variable can also be calculated separately
mean (data$sepal.width) SD (data$sepal.width)
Frequency tables and bar graphs for calculating categorical data species variables
table (data$species) Barplot (table (data$species))
For unary numeric data, it is common practice to draw histograms and box plots to observe their distributions:
hist (data$sepal.width) BoxPlot (data$sepal.width)
For two-dollar numeric data, you can observe the rule by scatter plots
plot (data$sepal.width,sepal.length)
If you need to save the drawing results, we recommend that you use the Plot menu command in Rstudio to select Save Plot as Image
Introduction to the basics of R language: Data import and description statistics