The Ultimate R Programming & Machine Learning Course

Download the base R from CRAN mirrors, install the 64-bit Windows version as administrator, choose a no-spaces path, and launch R with a frontend such as RStudio.

Explore the R console and RStudio interface, learn to run code, view history, manage environment and files, access help, and use packages and projects to organize work.

Learn how to set up and use R Studio, create and save files, initialize a git repository, commit changes, and push to GitHub or Bitbucket for version control.

Learn to install packages in R studio via the packages pane, including installing multiple packages and automatic dependency handling. Load packages with library or require, noting restarts reset loaded packages.

Learn to use R as a calculator, handling numeric, integer, character, and time-based data, storing variables, and applying functions with proper operator precedence and parentheses.

Learn how to assign and manage R variables using the left arrow and the equals sign, understand serial assignments, remove variables with rm, and recognize case sensitivity and tab completion.

Explore R data types, including numeric, integer, character, factor, date, date-time, and logical. See how R checks types, promotes numeric to decimals, and handles text and dates, with common pitfalls.

Learn to store data in vectors in R, create numeric vectors with c(), perform element-wise operations, generate sequences with colon, and use length and any and all checks.

Master how to store data in vectors, index elements with square brackets, name vectors, and work with factors and missing data in R, highlighting vectorized operations for efficiency.

Learn how R functions handle vectors, distinguish between vector-wide and element-wise operations, and specify arguments (positional or named) such as the mean function's trim and na.rm for missing values.

Explore advanced data structures in R, including vectors, arrays, matrices, and lists, and learn how data frames function like spreadsheet storage for efficient data analysis.

Create and explore data frames in R by combining numeric and character columns, name columns, prevent factor conversion with strings as factors, and use head and names to inspect data.

Learn to inspect and access data frames in R with head and tail, check type using class, and select rows, columns, or cells via brackets and the dollar sign.

Learn how to use R lists as versatile containers that hold numbers, vectors, data frames, and even nested lists, then create, name, subset, and extend them.

Create five-by-two and two-by-ten matrices, perform elementwise addition and multiplication, and compute the dot product after transpose, ensuring the matrices have matching dimensions and naming rows and columns.

Explore arrays in R, surpassing vectors and matrices with arbitrary dimensions. Learn indexing in multi-dimensional arrays using square brackets, selecting first rows and elements.

Learn how to read a CSV into R using read.table, specify header and comma separators, and optimize performance by using stringsAsFactors = FALSE.

Discover why pulling data from Excel is discouraged and adopt a practical workflow: convert Excel data to csv and read it into R, avoiding portability issues.

Learn how to read data from databases in R by connecting with a DSN and the Roadblock package, then query and join the orders and order_details tables using SQL.

Learn to use the foreign package to read data from SPSS and Stata files, navigate licensing hurdles, and access open source alternatives to load data from other tools.

learn to create, save, and load binary R data files that store one or more R objects, using save and load to preserve data frames across sessions.

Discover how to load data included with R packages using the data function, with examples like diamonds and tips, and how to list available datasets.

Learn to scrape web tables in R with the XML package by loading a URL, extracting a table into data frame, and setting header equals false and stringsAsFactors = false.

Learn to visualize data in R for exploration and reporting, mastering histograms, box plots, scatter plots, and line charts using the diamonds dataset to create compelling graphics.

Visualize a single continuous variable with histograms, showing carat frequency, and create scatterplots of price versus carat to reveal their relationship using the plot function and formula notation.

Draw a box plot for the carat variable with the boxplot function. Interpret quartiles, median, and 1.5 times the interquartile range, and note the box shows the middle 50 percent.

Explore ggplot2 for elegant statistical graphics in R, learning to initialize plots, map data with aes, and build layer-by-layer visuals using geoms like points, lines, or histograms.

Explore ggplot2 histograms and density plots with the diamonds data. Map x to carrot, add histogram and density layers, and adjust bin width or bandwidth; fill the density for clarity.

Explore building an attractive two dimensional scatter plot of diamond price versus carat, using layered ggplot, mapping color to diamond color and shape to clarity, with legends.

Learn to create box plots and violin plots in R, comparing diamond cuts, layering points and jitter for richer visuals, and controlling layer order for impact.

Create and refine multi-year line plots in R using ggplot, incorporating year and month variables, data subsetting, labeling, color grouping, and axis formatting for clear trend visualization.

Learn how to create small multiples to compare data across levels using ggplot2. Explore wrapping and faceting with grid and color aesthetics for diamonds data.

Learn how to map color, shape, and size in a Dewji plot scatterplot, with price on the axes and depth mapped to point size, for clear diamonds insights without overload.

Learn to apply flexible graph themes with Geoffrey Arnold's Themes package, loading pre-built or custom styles for Wall Street Journal and Economist aesthetics, colors, and legend placement.

Explore core programming basics by building a hello world function, assigning it to a variable, and calling it to print hello world, while learning about arguments, curly braces, and functions.

Explore how function arguments serve as inputs to control behavior, using positional and named arguments, default values, and a dot-dot-dot catch-all, illustrated with hello person examples.

Learn how to define R functions that return values from inputs, using last-line or explicit returns. Return various R objects such as numbers, strings, vectors, data frames, and lists.

Explore do.call to gain flexibility by supplying a function name or function object and a named arguments list for dynamic, higher-order calls, including applying mean.

Explore if statements to control program flow with equals, not equal, less than, less than or equal, greater than, greater than or equal operators, and consider whitespace.

Demonstrate staggered if statements with else by building a function that handles x equals one, zero, and other values, using print and comments to explain control flow.

Learn how to use the switch statement in R to test one argument against many results, with name-value pairs, a default option, and how numeric values map to positions.

Master vectorized checks in R using ifelse to test elements across full vectors, boosting performance. Note how missing data NA and returning a vector can yield character outputs.

Explore compound statements in R to test multiple conditions with vectors, using & vs && and | vs ||, including short-circuiting and nested checks for complex logic.

Explains how to use a for loop in R to iterate over vectors, print values, compute lengths, and assign results by indexing, with examples using numbers and fruit names.

Discover how to implement a while loop in R, starting at 1 and printing 1 through 5 as x increments.

Master R control loops by using break and next to skip iterations or terminate a for loop, illustrated with a 1 through 10 example.

Learn to manipulate data efficiently in R using apply and lapply for matrix operations, sum across rows or columns, and handle missing values with na.rm to speed up code.

Explore data munging with lists in R, applying functions across list elements using lapply and sapply to sum elements, convert results to named vectors, and understand list versus vector outputs.

Master data munging with mapply by iterating over two lists of matrices and a vector, comparing identical elements, and applying a custom function to add rows across lists.

Learn to use the aggregate function for data munging in R, computing mean price by cut and color with the diamonds dataset.

Discover data munging with the plyr package, cleaning baseball data and computing on-base percentage as a weighted average using the split-apply-combine paradigm.

Learn to combine datasets in R by stacking vectors into a matrix or data frame and by binding rows or columns using cbind and rbind, preserving column names.

Learn data munging by reading country codes and country data, then join them on a common key, using merge and join, and rename columns to align join keys.

Switch between wide and long data formats with melt and cast using reshape2 on the air quality dataset, using month and day as identifiers and metric as headers.

Explore how text data is becoming more pervasive in the world. Develop skills in combining text and ripping it apart, and examine extensive regular expression capabilities.

Explore text manipulation in R by combining strings with paste and sprint, learn how separators, vector recycling, and collapse affect output for building sentences from variables.

Extract and clean text from unstructured data using regular expressions and string operations in R, reading data from web sources, splitting and subsetting strings, and assembling tidy data frames.

Explore statistics basics including means, variances, correlations, and t tests, and learn the various ways to compute these measures using R.

Draw random numbers from normal, binomial, and Poisson distributions, examine densities and cumulative probabilities, and compute quantiles with pnorm and qnorm.

Explore core statistics in R: compute mean, variance, standard deviation, and correlations; handle missing values, use weighted averages, and summarize data with quantiles and heat maps.

Explore how to perform one-sample, two-sample, and paired t-tests and ANOVA in R, assess normality and variance with Shapiro tests and variance checks, and visualize results with ggplot.

Explore regression as the workhorse of statistics, covering linear models and generalized linear models, including logistic, Poisson, and survival models, and learn to fit them in R. Assess model performance with mean squared error and deviance, interpret coefficients and residuals, and compare via R-squared, using father–son height data and tips data to illustrate simple and multiple regression.

Explore extending simple regression to multiple regression using NYC housing data. Visualize value per square foot with histograms and scatterplots, and compare boroughs through small multiples.

Fit regression models to predict value per square foot using units, square footage, and borough with dummy variables and interactions. Interpret coefficients, compare models, and predict with intervals in R.

Familiarize with generalized linear models by fitting a logistic regression on ACS data to predict income above 150000, using logit link, inverse logit, and deviance.

Learn how Poisson regression models count data with a log link and how to diagnose and address over-dispersion using quasi Poisson or negative binomial models.

Explore survival analysis in r using the survival package, fit cox models for censored data, and interpret survival curves and stratified hazards across patient groups.

Assess model quality using residuals and diagnostic plots, including residuals vs fitted, QQ plots, and residual histograms, with fortified data and cross-validation as modern alternatives.

Compare multiple models on housing data to identify strong predictors, using coefficients, RSS, AIC, BIC, and deviance to guard against overfitting, and extend to generalized linear models including logistic regression.

Explore k-fold cross-validation for linear models, using the boot package to compare five housing models with mean squared error and adjusted cross-validated error to identify the best performing model.

Learn how to use the bootstrap to estimate uncertainty and build a 95% confidence interval for batting averages with baseball data in R.

Explore stepwise variable selection for linear models, using a null model and a full model with a bidirectional search, guided by information criterion (aic) to choose predictors.

Explore regularisation in high-dimensional data with the elastic net, combining ridge and lasso to prevent overfitting, select variables, and use cross-validated lambda and alpha to shape coefficients.

Use bayesian shrinkage with weakly informative priors to tame out-of-control coefficients in yearly voter ideology models, reducing uncertainty and improving interpretation over time.

Use nonlinear least squares when a linear model won't fit to locate a wifi hotspot from x y coordinates and distance, estimating beta_x and beta_y in R.

Explore smoothing splines and natural cubic splines to fit curved relationships in data, using diamonds dataset, varying degrees of freedom, and visualizing with ggplot2, and compare straight vs curved fits.

Explore generalized additive models to capture non-linear relationships by combining variables with smooth functions and tensor products, then fit a binomial family regression for credit data.

Explore how decision trees handle non-linear data for regression and classification, including visualizing splits. Learn how random forests boost predictions by averaging many trees on credit data.

Learn how autocorrelation affects time series data and apply AR and MA modeling, using ACF and PACF to identify structure, with World Bank GDP data as a practical example.

Explore how ARIMA models combine autoregressive, integrated, and moving average components, use auto daydreamer to select orders, and forecast with predict and forecast while inspecting residuals with ACF.

Explore multivariate time series modeling with vector autoregression (VAR) on GDP data, transforming data to wide format, selecting lags, and forecasting with confidence intervals across countries.

Learn how to use garch models for volatility in financial time series, demonstrated with AT&T stock data, specification, fitting, diagnostics, and log returns comparison for improved volatility modeling.