Most of the times the explanatory variables are not linearly related to the response variable and we need to find the best model for our data. In this type of situations, we move on to polynomial models to check whether they will be helpful in determining the accuracy of the predictions. This can be done by using power of the independent variables in lm function.ExampleConsider the below data frame −> set.seed(99) > x1 x2 x3 x4 y df PolynomialModel1 summary(PolynomialModel1) Call: lm(formula = y ~ x1 + I(x1^2) + x2 + x3 + x4) Residuals: Min 1Q Median 3Q Max ... Read More

Since no one is perfect, people might forget to add all columns that are necessary for the analysis but this problem can be solved. If a column is missing in our data frame and we came to know about it later then it can be added easily with the help of reordering the columns.ExampleConsider the below data frame −> x1 x2 x3 df df x1 x2 x3 1 1 a 1 2 2 b 2 3 3 c 1 4 4 d 2 5 5 e 1 ... Read More

While doing the analysis, we might come across with data that is not required and we want to delete it. This data can be a whole row or multiple rows. For example, if a row contains values greater than, less than or equal to a certain threshold then it might not be needed, therefore we can delete it. In R, we achieve this with the help of subsetting through single square brackets.ExampleConsider the below data frame −> set.seed(99) > x1 x2 x3 x4 x5 df df ... Read More

Sometimes when we read data in R, the missing values are recorded as blank spaces and it is difficult to replace them with any value. The reason behind this is we need to know how many spaces we have used in place of missing values. If we know that then assigning any value becomes easy.ExampleConsider the below data frame of vectors x and y.> x y df df x y 1 1 2 3 2 3 2 4 1 43 5 2 2 6 3 7 2 3 ... Read More

Correlation matrix helps us to determine the direction and strength of linear relationship among multiple variables at a time. Therefore, it becomes easy to decide which variables should be used in the linear model and which ones could be dropped. We can find the correlation matrix by simply using cor function with data frame name.ExampleConsider the below data frame of continuous variable −> set.seed(9) > x1 x2 x3 x4 x5 df df x1 x2 ... Read More

Ordering columns might be required when we want to manipulate the data. Manipulation can have several reasons such as cross verification, visualisation, etc. We should also be careful when we change anything in the original data because that might affect our processing. To change the order of columns we can use the single square brackets.ExampleConsider the below data frame −> set.seed(1) > Class Grade Score df df   Class Grade Score 1   a     A     68 2   b     B     39 3   c     C      1 4   ... Read More

There are different ways to express any chart. The more information we can provide in a chart, the better it is because a picture says thousand words. Since nobody likes to read a long-reports, we should have better reporting of charts. Therefore, we can add a chart title as well as chart sub-title in ggplot2 to help the readers.ExampleConsider the below data −> set.seed(1) > x table(x) x 2 3 4 5 6 7 8 9 11 1 3 4 2 4 2 2 1 1 > df library(ggplot2)Creating a simple bar chart −> ggplot(df, aes(x))+ + geom_bar()OutputCreating a ... Read More

There are times when duplicated rows in a data frame are required, mainly they are used to extend the data size instead of collecting the raw data. This saves our time but surely it will have some biasedness, which is not recommended. Even though it is not recommended but sometimes it becomes necessary, for example, if it is impossible to collect raw data then we can do it. If we do so then we must specify it in our analysis report. In R, we can use rep function with seq_len and nrows to create a data frame with repeated rows.ExampleConsider ... Read More

Sometimes subsetting of group wise maximum values is required while doing the data analysis and this subset of the data frame is used for comparative analysis. The main objective is to compare these maximums with each other or with a threshold value. In R, we can find the group wise maximum value by using group_by and slice functions in dplyr package.ExampleConsider the below data frame −> x y df head(df, 20) x y 1 S1 1 2 S1 2 3 S1 3 4 S1 4 5 ... Read More

The warning “removed n rows containing missing values” occurs when we incorrectly specify the range of the values for X-axis or Y-axis. We can this range in ggplot function using scale_x_continuous(limits=c(?, ?)) for x axis and scale_y_continuous(limits=c(?, ?)) for y axis. If the range will be larger than the actual data range then there will be no warning otherwise, we will get the warning for the number of missing values.ExampleConsider the below data frame −> set.seed(2) > x y df library(ggplot2)Creating the plot with Y-axis limits from 0 to 5−> ggplot(df, aes(x, y))+ + geom_point()+ + scale_y_continuous(limits=c(0, 5)) Warning message: ... Read More