By default, the shape of legend is circular but we can change it by using the guides function of ggplot2 package. For example, if we have a data frame with two numerical columns say x and y, and one categorical column Group then the scatterplot between x and y for different color values of categories in categorical column Group having different shape of legends can be created by using the below command −ggplot(df, aes(x, y, color=Group))+geom_point()+guides(colour=guide_legend(override.aes=list(shape=0)))Here, we can change the shape argument value to any value between starting from 0 to 25.Consider the below data frame −Example Live DemoxRead More

When we do subsetting with the help of single square brackets we need to be careful about putting the commas at appropriate places. If we want to subset rows using the columns then comma needs to be placed before the condition. The “undefined columns selected” error occurs when we do not specify any comma. Check out the examples to understand how it works.Consider the below data frame −Example Live Demox15),]Output   x1 x2 1  7  0 2  6  4 4  6  1 7  6  1 9  7  3 11 6  3 12 9  2 15 7  4 16 7  3 17 6  2 18 6  3Example Live Demoy1

To collapse data frame rows by summing using dplyr package, we can use summarise_all function of dplyr package. For example, if we have a data frame called df that has a categorical column say Group and one numerical column then collapsing of rows by summing can be done by using the command −df%>%group_by(Group)%>%summarise_all(funs(sum))Consider the below data frame −Example Live DemoGroup

Subsetting is one of the most important aspects of data analysis. One such situation could be subsetting the character column based on multiple values. For example, if a character column of an R data frame has 5 categories then we might want to extract only 2 or 3 or 4 values then it can be done by using the filter function of dplyr package with str_detect function of stringr package.Consider the below data frame −Example Live DemoGroup

If a vector value exists in another vector then we might want to find the frequency/count for such values in the other vector. For example, if we have two vectors say x and y, and some of the values in y exists in x as well. Therefore, we can find the frequency of values in x for y values can be found by using the command colSums(outer(x,y,"==")).Example Live Demox1

If we have time series data stored in a data frame then plotting the same as a time series cannot be done directly, also the labels for the series might not be possible directly. Therefore, we first need to convert the data frame to a time series object by using the function ts as shown in the below example and then using the plot function to create the plot, this will display the labels for the series as well.Consider the below data frame −Example Live DemoTime

By subtotal we mean finding the sum of values based on grouping column. For example, if we have a data frame called df that contains three numerical columns as x, y, z and one categorical column say Group then the subtotal of x, y, z for each category in Group can be found by using the command aggregate(cbind(x,y,z)~Group,data=df,FUN=sum).Consider the below data frame −Example Live Demox1

To create a random vector of integers with increasing values, we can do random sampling with sample.int and for increasing values cummax function needs to be used. For example, to create a random vector of integers of size 5 up to values 5 starting from 1 can be done by using the command cummax(sample.int(5)).Example Live Demox1

When we perform any type of data analysis, there are many types of objects that are created in the R environment such as vector, data frame, matrix, lists, arrays, etc. If we want to get the list of available data frames in R environment then we can use the below command −names(which(unlist(eapply(.GlobalEnv,is.data.frame))))Example Live Demox1

If we have a numeric column in an R data frame and the unique number of values in the column is low that means the numerical column can be treated as a factor. Therefore, we can convert numeric columns to factor. To do this using dplyr package, we can use mutate_if function of dplyr package.Loading dplyr package and converting numerical columns in BOD data set (available in base R) to factor columns −Examplelibrary(dplyr) str(BOD) 'data.frame': 6 obs. of 2 variables: $ Time : num 1 2 3 4 5 7 $ demand: num 8.3 10.3 19 16 15.6 19.8 - ... Read More