Introduction to R Programming

Introduction to R

• R is a calculator
• R can evaluate complex mathematical expressions
• Variables are assigned using the <- operator (e.g., x <- 10)
• Basic arithmetic functions are available (e.g., 1+1, sqrt())
• Functions for creating sequences (seq(), rep())
• Functions for calculating absolute values (abs())
• Functions for manipulating decimal places (e2, e-2)
• Element-wise product can be performed with ab; uv

Functions in R

• R has built-in functions for performing various tasks, including mathematical calculations.
• Functions can be used to perform simple arithmetic operations or complex analyses.

Reading Files in R

• read.delim() is used for tab separated files (.txt)
• Default decimal separator is "."
• read.table() reads files in tabular format to create a data frame
• read.csv() reads comma separated values files (.csv) into a data frame.
• read.csv2() is used when the decimal separator is "," and the field separator is ";".

Exploring Data

• View(x) displays the data frame's contents.
• head(x) shows the top 6 rows, head(x, n=n) for the first few rows
• tail(x) displays the last 6 rows, tail(x,n=n) for the last few rows
• names(x) shows the names of the variables in the data frame.

Types of data in R

• R supports several data types including numeric, character and logical, vectors, matrices and dataframes

Data Types

• Scalars: Single value (e.g., a <- 5)
• Vectors: Multiple values (e.g., v <- c(1,2,3))
• Matrices: Two-dimensional array of values (e.g., m <- matrix(v, 3, 2))
• Lists: Can contain varied data types (e.g., q <- list(a=v, b=x, c=u))
• Data frames: Table-like structure, typically used for storing data with different variable types.

Data Structures

• Vectors store multiple values of the same type.
• Matrices are two-dimensional structures.
• Lists can contain elements of different data types.
• Data frames are tabular structures, organized into rows and columns.

Operations (Arithmetic)

• 3 + exp(4) * 2^2
• (3 + exp(4)) * 2^2
• R follows operator precedence rules when evaluating expressions.

Data Frames

• A table-like structure with different variable types.

Data Merging

• cbind() merges by columns.
• rbind() merges by rows.
• merge() joins two datasets based on a common variable. (Commonly used for merging two datasets).

Data Manipulation

• na.omit() removes rows with NA (missing) values.
• complete.cases() filters out rows with missing values.
• apply, colMeans, rowMeans, mean, used to calculate and analyze data.
• summary() provides summary statistics for data frames or vectors.

Functions Creation

• User-defined functions are created in R using the function() syntax.

Data Types in R (continued)

• Character data: Text data (e.g., name <- c("Ahmed", "Laila")).
• Logical data: TRUE/FALSE values (e.g., smoker <- c(TRUE, FALSE, FALSE)).
• Numeric data stores numbers.
• Ordering variables: sort(c(4,2,6))
• colnames(m)=paste("X",1:ncol(m),sep="") renames the column names.
• rownames(m)=1:nrow(m) renames the row names.

Data Preparation

• Data import from different formats: read.csv, read.table.

Missing Values Analysis

• is.na(dataNA) identifies missing values.
• sum(is.na(data), apply(data, 2, sum) gives counts or the total of missing data by column.
• Common functions include na.omit(), complete.cases(), colSums(is.na(data)) to find the missing values in a column/data.
• Handling missing values, including replacing them with imputation methods (mean, median etc).

Outliers

• Identifying: boxplot(data).
• Removal: data[data > bench]
• Filtering data using calculated quartiles (Q1, Q3, IQR).

Descriptive Statistics

• Calculating measures like mean, median, min, max, range, IQR, standard deviation, variance.
• Using functions such as mean(), median(), min(), max(), range(), IQR(), sd(), var().
• Determining the mode: Mode() calculated using the DescTools library.
• summary() summarises numeric, logical and/or factor.

Tables and Plots

• Creating frequency tables, contingency tables (crosstabulations)
• Creating histograms to plot data (including specific quantitative or qualitative variable(s)).
• Constructing boxplots to identify data distribution.
• Using table(), chisq.test(), fisher.test(), oddsratio(), assocstats(), and cor().

Association and Correlation

• Calculate and display correlations of variables.
• Create contingency tables for association analysis - using measures like odds ratio, relative risk or chi-squared test vcd(), vcdExtra and nnet libraries.

Data Manipulation & Visualisation

• Data manipulation using data.frame(), cbind(), rbind(), dplyr.
• Scatter plots: The plot() function can be used.

Additional Considerations:

• Load the necessary libraries at the start of your script: using library command.
• Using attach() to make variables in a data frame directly accessible.
• Appropriately handle data types when using specific functions (e.g., converting factors to numerical values with as.numeric()).