Http://www.cnblogs.com/wentingtu/archive/2012/03/03/2377965.html
Summarize the available R language resources on the Internet http://www.douban.com/note/262946592/?type=like R language: Machine Learning packages
Http://blog.sciencenet.cn/blog-634847-497887.html
Reproduced R Basics: R language and its learning resources
Http://www.biosino.org/R/R-doc/onepage/R-data_cn.html#Importing-from-other-statistical-systems
Import and export of R data
Http://www.biostatistic.net/thread-40320-1-1.html
READ.DBF (Foreign)
We can see what variables are in work space through LS ().
Help (functionname) functionname args (functionname) example (functionname)
Use Object.size () to see how much memory each variable occupies.
View current work space's memory usage through memory.size ()
Use the Memory.limit () to see the maximum memory usage limits specified by the system. If you think the current memory limit is insufficient, you can change to a new upper limit by Memory.limit (Newlimit).
Note that in 32-bit R, the cap is capped at 4G, and you cannot use more than 4G (upper limit) on a program. At such times, you might consider using a 64-bit version. For some large but useless intermediate variables, get into the habit of cleaning up: You can use RM (object) To delete variables, but remember, RM remember to use GC () to do garbage collection, otherwise the memory will not be automatically released, the equivalent of you did not do RM.
Using the foreign package of READ.DBF () to read hundreds of DBF files can not
Http://blog.sina.com.cn/s/blog_62b37bfe0101f4h0.html
Accessing a column of data
For removing a row of data, you can use the subset () function or subscript operations, which can be vectors, matrices, and data frames. Taking the matrix as an example, when using subscript for data deletion, the deletion of whole rows or columns is often done, and by default, columns are deleted.
> X<-data.frame (Matrix (1:30,nrow=5,byrow=t))
>new.x1<-x[-c (1,4)] #去除第一和第四列
>new.x1<-x[-c (1,4),] #去除第一和第四行
> New.x1<-x[,-c (1,4)] # Remove first and fourth columns
Using the subset () function to access and select data frame data is more flexible, the subset function returns a subset of vectors, matrices, and data frames that meet the criteria.
Three ways to apply the subset function:
Subset (x, subset, ...)
Subset (x, subset, select, Drop =false, ...) # #对于矩阵
Subset (x, subset, select, Drop =false, ...) # #对于数据框
X is an object, subset is a logical expression that retains elements or rows, and Na is substituted for missing values.
Select is the selected range and should be less than X.
>x<-data.frame (Matrix (1:30,nrow=5,byrow=t))
> Rownames (x) =c ("One", "I", "three", "four", "five")
>colnames (x) =c ("A", "B", "C", "D", "E", "F")
> x
>new<-subset (X,A>=14,SELECT=A:F)
>new # # Select the A>14 row from a to f column.
http://my.oschina.net/zarger/blog/102818
access a column of data
The 1.1 str function can view the properties of each variable in the data frame:
> str (SQUID)
Data parameters in 1.2 functions--best way to access variables in a data frame
> M1 <-lm (GSI ~ factor (location) +factor (year), data = Squid)
1.3 $ symbol Another way to access a variable
> Squid$gsi or > squid[,6]
1.4 The Attach function to add a data frame to the search path of R, you can view the GSI data directly from the GSI command
>attach (Squid)
> GSI
Filter Data
Data2 <-data[data$v6==2,]
Y <-x[(X$timeindex >) & (X$timeindex <5000),]
X3 <-x2[(x2$date< as. Date ("2015/1/5")),]
X2 <-X1[order (x1[,3]),]
r for data filtering
http://blog.163.com/xiaoji0106@126/blog/static/136134661201392532614464/
References to box columns :
DN, D[,n], D "Name", d[, "name"], D$name gets the vector of a column.
D[n], d["name"] gets the data frame of a column.
D[c (M,n,...)], d[,c (M,n,...), D[,c ("name1", "name2",...)] Get a data frame of several columns
Other tips: Negative sign for culling.
You can use grep () to search for variable names. For example Mydata[grep ("^q", Names (MyData))] Select the data column whose name begins with "Q".
Reference to the data frame row :
D[n,] Gets a data frame that consists of a row.
D[c (M,n,...), n] gets a data frame consisting of several rows.
Head () gets a data frame consisting of the first 6 rows.
Tail () Gets the last 6 rows of the data frame.
Http://www.cnblogs.com/youxilua/archive/2012/01/12/2320455.html the more advanced
Http://www.360doc.com/content/13/1221/22/7440765_339121786.shtml Low to Advanced
http://blog.sina.com.cn/s/blog_5de124240101q5vw.html R Language Drawing progressive
Http://blog.sina.com.cn/s/blog_6cfc336b01018wcg.html
Http://www.cnblogs.com/holbrook/archive/2013/05/13/3075777.html
Drawing
Output Data
(1) R for Beginners (Chinese version). pdf
Write.table (x1,file= "E:\\siemens\\trafficdata\\result (). csv", append=false,quote=false,sep= ",", eol= "\ n", na= " NULL ", dec=". ", Row.names=false,col.names=true)
(2)
Cat ()
Http://blog.sina.com.cn/s/blog_5de124240101pwyv.html
Plot parameter
Http://blog.sina.com.cn/s/blog_6a02b6330101abn5.html
Plot parameter row multiple graphs
Par (mfrow=c (2,3)) a chart shows 2 rows, 3 columns. Mfrow Mfcol
Http://www.biostatistic.net/thread-94936-1-1.html
R-Language graphics Interactive iplots graphic interaction package
Http://www.cnblogs.com/speeding/p/4060500.html
# date to convert
Fitbit$date <-as. Date (Fitbit$date, "%y year%m month%d Day")
Strptime ("", format)
typeof (Sys.date ())
Http://www.biostatistic.net/thread-7035-1-1.html
R language: As. Posixlt () function Chinese help document (bilingual) Date- time conversion function
http://www.cnblogs.com/speeding/p/4159264.html
Solutions for Rjava packages that cannot be loaded in the R locale
Run the following command in the R language:
Sys.setenv (java_home= ' C:/Program Files (x86)/java/jdk1.7.0_55/jre ')
Sys.setenv (java_home= ' D:/programfiles/java/jdk1.7.0_40/jre ')
or add it in the profile R\r-3.1.2\etc\rprofile.site
The 32bits R must be 32bits java.
http://blog.fens.me/r-rjava-java/
http://www.haodaima.net/art/2522754
JAVA calls R language
1. Install Rjava-install.packages ("Rjava")-/rjava/jri/there will be 3 Jri files.
2. System environment variable settings: path add
... \library\rjava\jri
... R\win-library\3.1\rjava\jri\x64
... \r\r-3.1.2\bin\x64
3. Import 3 Jri files in a Java project.
4. Eclipse Run Settings Add VM arguments:-djava.library.path= "C:\Users\zhangjiajie\Documents\R\win-library\3.1\rJava\jri\x64 "
Rjava Tomcat Configuration
1) Three jar packages are placed in the Lib directory;
2) \rjava\jri\i386\jri.dll placed in the \tomcat8.0\bin\ directory;
Fitting
Linear regression
M <-LM (Y~x1+x2+x3,data=dataframe)//M saves the regression model
ANOVA (m)//Variance Analysis Table
Coef (m)//model factor
Confint (m)//confidence interval of regression coefficients
Deviance (m)//residuals squared sum
Effects (m)//orthogonal effect vectors
Fitted (m)//fit vector of y values
Residuals (m)//model residuals
Resid (m)//model residuals
Summary (m)//r2,f statistics
Vcov (m)//covariance matrix
Linear regression with interacting items
Lm (y~x1*x2), y = a*x1+b*x2+c*x1*x2+d
Lm (Y~X1*X2*X3)
Y =a*x1+b*x2+c*x3+d*x1*x2+e*x1*x2+f*x2*x3+g*x1*x2*x3+h
Lm (Y~X1+X2+X3+X1:X2:X3)
y = a*x1+b*x2+c*x3+d*x1*x2*x3+e
SETP stepwise regression, you can remove the meaningless variable backwards, you can add a new variable to the forward regression
Lm (y~x1, subset=1:100) selects only the first 100 data for regression
Lm (Y~i (X1+X2)) to (X1+X2) regression
Lm (Y~ploy (x,3,raw=true)) Y is the three-quadratic polynomial regression of x
Lm (log (y) ~ x1)
System.time (r_expression)//R run time
Calculate correlation coefficients
Http://www.biostatistic.net/thread-7287-1-1.htmlby
By (data, INDICES, fun, ..., simplify = TRUE)
Http://www.biostatistic.net/thread-7388-1-1.htmlcor
Cor (data)
By (data, data$x, Cor)
Http://www.klshu.com/tag/r%E8%AF%AD%E8%A8%80/page/2
Happy Tree R language
Http://www.klshu.com/1798.html
Http://www.klshu.com/1077.html using R to complete decision tree classification
Http://www.klshu.com/1555.html
Ggplot
http://www.klshu.com/1591.html Plot and Ggplot2 drawing comparison
Http://www.klshu.com/1667.html
SVM pattern Recognition, classification, regression analysis
Http://www.klshu.com/1430.html
Apply function Family
Http://www.klshu.com/1202.html
Http://www.klshu.com/1175.html
Http://www.klshu.com/1188.html
Http://www.klshu.com/1107.html
Association Rules
Http://www.klshu.com/1185.html
Common functions
Http://www.klshu.com/1144.html
Http://www.klshu.com/1147.html R language with shiny package to quickly build interactive Web applications
Http://www.klshu.com/1073.html
Set of data mining functions
the finding of http://www.klshu.com/1071.html K-means clustering number
Http://www.klshu.com/238.html
Http://www.klshu.com/693.html
Http://www.klshu.com/719.html
Http://www.klshu.com/107.html Basic
Drawing
Http://www.klshu.com/25.html
Three-dimensional drawing
Http://www.klshu.com/21.html
R Code Specification
Http://www.mamicode.com/info-detail-374357.html
R Language and Data Analysis VI: A brief introduction to time series
Http://www.empowerstats.com/cn/download.html
Download Easy Empower Statistics
http://yanping.me/shiny-tutorial/
Chinese tutorial: Building shiny Applications with R
Http://book.51cto.com/art/201408/449427.htm
Xts
http://blog.csdn.net/desilting/article/details/39013825
ARIMA (P,D,Q) model, ACF, PACF
Http://www.biostatistic.net/thread-6683-1-1.html
AIC Information Guidelines
BIC Information Guidelines
Http://www.biostatistic.net/thread-40266-1-1.html
Forecast package. Auto.arima (): Returns the best Arima model based on the AIC,AICC or BIC value
diff (x,lag=n) #滞后差分, lag is used to specify several lags. The default lag value is 1.
X<-c (1,5,23,29)
diff (x, lag=n) = = = C (x[1+n]-x[1], x[2+n]-x[2], x[3+n]-x[3],...)
http://blog.163.com/zzz216@yeah/blog/static/16255468420147179438149/
DPLYR Data Processing Package
http://ju.outofmemory.cn/entry/84555
Processing time data with lubridate packets
return
Http://www.cnblogs.com/luosha/archive/2012/06/30/2571542.html
Linear regression LM and prediction
http://www.douban.com/note/298285612/
LM Results Analysis
Http://www.biostatistic.net/thread-7433-1-1.html
Linear regression LM
Summary looks at the results returned by LM.
In residuals: we can see some information about the residuals: minimum maximum, 4-cent, etc. Coefficients: Medium is the most critical of the relevant estimates for C and B. Where estimate is the estimate with the B,c value, Std. Error is the standard deviation of the regression parameters B and c: SD (b), SD (c). The remaining two parameters are hypothesis tests of the regression parameters: T value is the t of the hypothesis test for B,c, and the P-value (used to compare with the significance level to decide whether to accept the AH hypothesis test) Pr (>|t|).
Finally we can see 3 * numbers, which indicates that X