Intro to R software:
>data entry
bt=c(1,2,3,4,5,5,4,2,3,1)
>length(nameofdata)
this command gives 'n' or 'number of observations
>nrow(nameofdatatable)
-to check number of rows in data table
>ncol(nameofdatatable)
-to check number of column in data table (data.frame)
>dataname[,-1]
-this command is used to remove column
>dataname[c(2,17,8),]
-this command is used to locate multiple rows data
>dataname[,c(6,4,12)]
-this command is used to locate different column data
>dataname[1:10,]
-this command gives rows from 1 to 10
>dataname[,1:10]
-this command gives coloumn from 1 to 10
>data.frame(nameofdata,variables)
this command gives data in tabular form
>nameofdata[,1]
-this command locate column in give data table
>mean(nameofdata)
to find mean of sindle variable,data
>mean(nameofdatatable[,2])
another method of calculating mean
>mean(nameofdatatable[,2][1:50])
-to calculate mean for column 2 from entries 1 to 50
>apply(datatablename,2,mean), same for median,sd,var
to find collective mean of data
2 position for coloumn
1 position for row
>aggregate(datatablename[,5]~datatablename[,3],data=datatablename,FUN=mean)
-this commands gives mean against each category in datatable
-place numeric column first then category column later
-~ 'tilta' sign is used to link one column to another
>aggregate(datatablename[,5]~datatablename[,3],data=datatablename,FUN=sd)
-to get sd against each category
>aggregate(datatablename[,5]~datatablename[,3],data=datatablename,FUN=summary)
-to get summary of these column
>sort(dataname)
to arrange data in ascending order
>unique(dataname)
to exclude repitition of same values
>plot(dataname)
graphical representation of data in Scatter plot
>plot(dataname,col="green")
to add colors in graph
>plot(dataname,col="red",xlab="bodytemperature",ylab="weights")
to add names on graph
>plot(dataname,col="red",xlab="bodytemperature",ylab="weights",main="graph of bt and weights")
to add name of graph
>boxplot(dataname,col="red",xlab="bt",ylab="weights",main="graph")
this command gives boxplot of data
>summary(nameofdataordatatable)
this commands gives information and range of data or quartiles
category and length of datatable
>apply(nameofdata,2,mean,na.rm=T)
if we have different lengths of data then we add "NA" in missing
positions to equal lengths of data
then in computing mean add "na.rm=T" command to remove NA during calculation
>data()
this commands gives us datasets already available in R software
>head(nameofdata)
this commands gives us some first values from data
>tail(nameofdata)
this command gives us some last values from data
>str(nameofdata)
-this command gives structure of dataset
-like number of observations/ number of variables
>getwd()
this command gives the directory of R
>setwd("D:/Nabeela/mphil 2 semester/stat/data sets")
this commands enters the directory where datasets are saved
>namethedata=read.csv("D:/Nabeela/mphil 2 semester/stat/data sets/iq_level.csv")
-this command read the file in drive saved in excel form
-but save this excel file in File Format CSV(Comma delimited)
>namethedata=read.table("D:/Nabeela/mphil 2 semester/stat/data sets/quail_partial_data.txt")
-this command import data file in R saved in notepad form
>namethedata=read.table("D:/Nabeela/mphil 2 semester/stat/data
sets/quail_partial_data.txt",header=T)
-this command removes header from the table in saved file
>windows()
-to have more windows for graphs on other windows
>par(mfrow=c(2,5))
-to divide windows to get more than one
>attach(dataname)
-this command is used to get any column from given data table
-after applying this command just type name of column you want to work
on and click enter
>nameofdata[,1]
-same as above
>nameofdata$nameofcolumn
-same as above
>cbind(nameofcolumn,nameofothercolumn)
-this command is used to bind two columnes of different datasets
>rbind(nameofcolumn,nameofothercolumn)
-this command is used to bind two coloumns and represent in row format
>table(dataname[,2])
-this command gives frequency of values in data
> plot(iris[,1],col="black",ylim=c(0,8))
> points(iris[,2],col="red")
> points(iris[,3],col="blue")
> points(iris[,4],col="green")
-to get plot and add further points on it
>givename=which(variablename=="valueofvariable")
-it split vaiable column with same values
-f=which(Maternal=="F1") OR p=Maternal[-f]
-> f
[1] 1 2 3 4 22 23 24 25 26 27 28 29 30 31 32 47 48 67 68
[20] 69 70 71 72 73 74 75 76 77 78 79 80 97 98 99 100 101 102 122
[39] 123 124 125 126 127 128 129 130 144 145 146 147 148 169 170 171 172 173 174
[58] 175 176 177 178 179 180 181 182
-> Maternal[f]
[1] "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1"
[16] "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1"
[31] "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1"
[46] "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1" "F1"
[61] "F1" "F1" "F1" "F1" "F1"
REGRESSION MODEL (SIMPLE OR LINEAR)
STEPS:
1-TO CHECK LINEARITY PLOT SCATTER PLOT BETWEEN DEPENDENT VARIABLE (Y) AND INDEPENDENT
VARIABEL (X)
2-DRAW HISTOGRAM TO CHECK NORMALITY.
3-THEN APPLY LINEAR MODEL FUNCTION IN R.
4-INTERPRET RESULTS BY TAKING SUMMARY OF Lm.
>lm(dependentvariable~independentvariable)
-this commands is for linear model function
>plot(independentvar,dependentvar)
-to get scatter plot for regresion model
>points(independentvar,yhat,col="anycolor")
-to mark points for line of best fit or regression line
>lines(independentvar,yhat,col="anycolor")
-to connect points through lines
>abline(lm(dependentvar~independentvar),col="red")
-to get regression line in scattor plot
>pred=predict(fit,newdata=nameofnewdata)
-to get prediction of new data set based on previous data lm results
>cor(nameofdata)
-to get corelation between variables
>givename=which(nameofcolumn=="nameofcategory")
-for example: Gender=which(Sex=="M")
>givename
-for example:
Gender
1 3 7 9 13 16 17 18 19 20 21 23 24
25 27 30 31 32 33 34 40 41 42 43 48 49 50 51 52 58
-this gives positions where M category is placed
>nameofcolumn[namegiven]
-for example: Gender[male]
"M" "M" "M" "M" "M" "M" "M" "M" "M" "M" "M" "M" "M" "M" "M"
"M" "M" "M" "M" "M" "M" "M" "M" "M" "M" "M" "M" "M" "M" "M"
-this commands give value of given data name
>cor(nameofdata[,-c(1:3)])
-this commands compute corelation but removing 1 to 3 column
-Ho=no relation ;H1=relation
> library(nameofpackage)
-to get package of test you want to perform
-for example
library(ppcor)
>pcor(nameofdata)
-to perform partial corelation
>cor(variableone,variable2,method="spearmen")
-to get rank corelation with spearmen method
>cor(variableone,variabletwo,method="kendal")
-to get rank corelation with kendal method
>cor.test(variableone,variabletwo,method="kendal")
-to get colrelation with p value
>cor(nameofdatatabel,method="spearman")
-to get table of spearman corelation analysis
>ad.test(nameofdata)
-to get anderson darling test for normality
-Ho=normal ;H1=not normal
>shapiro.test(nameofdata)
-to perform normality of test
-Ho=normal ;H1=not normal
>
testing of hypothesis:
TESTING OF HYPOTHESIS:
STEPS:
1:Normality
2:Homogenity
3:tests
normal and homogenous t-test, var true
normal and non homogenous t-test, var false
non normal wilcox test
Paired data t-test paired ture
#########################################AFTER MIDS ###########################
>dataname=rep(1:4,each=5)
-this command is used to repeat 1 to 4 counting 5 times
-for example treatment= (1,1,1,1,1,2,2,2,2,2,3,3,3,3,3,4,4,4,4,4)
>overallmean=mean(name of data $ treatment output name )
-to caculate overall mean
>txmean=tapply(name of data $ treatment output name , name of data $ treatment name ,mean)
-to calculate means of treatments individually
>duncanTest(fitt)
-to apply duncan test
>tukeyHSD(fitt)
-to apply tuckey honest significant difference test
