R: Complete Data Analysis Solutions

Learn to rename data variables in R by using download.file to fetch datasets from GitHub, loading them with read.csv, and updating column names with names, call names, and dimnames.

Explore converting data types during preprocessing by using class and tr functions to inspect types, convert birth date and hire date to date, and set names to character.

Learn to manipulate date data with a date package, convert data types, rename columns, and calculate employee age by computing intervals between birth date and hire date, using year-based results.

Use the bind function to add new records and attributes in the employees data set, reassign results, and compute age, position, and marital variables with transform.

Learn to subset and filter data with square brackets, head and tail, and the subset function. Apply conditions to select rows and columns, including salary ranges and gender.

Drop data by excluding rows or columns with negative indices and the within function to remove unwanted attributes, discarding bad data during pre-processing.

Merge datasets by a common key with the merge function and plyr join, enabling left or right joins, then sort by salary and from date using sort and order.

Reshape data in R using diecast to convert long to wide and melt to revert, applying shaping formulas with employee number and year. Remove missing values with na.omit.

Enhance a data frame with a data.table to accelerate aggregation and descriptive statistics for large datasets, and compare performance with traditional data frames.

Learn fast aggregation with data.table in R, calculating average prices and totals by date, counting users, sorting results, and creating new columns for merged data sets.

Merge large datasets with data.table by performing inner, left, right, and full outer joins on product, then use set key to sort and summarize prices by product.

Learn to subset and slice data with deployer, using filter, slice, and head on data frames and data tables, including multiple conditions with in and or operators.

Use sample_n to randomly slice rows and sample_frac to select a percentage from the original dataset. Learn to specify the weight parameter whose length matches the number of rows.

Learn to select specific columns with the dplyr select function, including quantity and price, exclude with minus, and use patterns such as starting with P.

Learn to chain dplyr operations with the chain operator, selecting columns, filtering data, using head, and applying sum, while noting the deprecated dot operator in favor of the greater-than operator.

Master summarizing data with dplyr using group by, summarize, and summarize_each to compute total purchases, min and max spend per user, and N_distinct counts across data frames and tables.

Learn how to merge data with dplyr by applying inner, left, right, full, and anti joins to combine datasets and retrieve nonmatching rows.

Generate random samples from a population using R functions, control reproducibility with seed settings, and simulate experiments like lotteries, coin tosses, and dice to illustrate probability, distribution, and statistical inference.

Explore generating samples from a uniform distribution, visualize with histograms and bar plots, and simulate a fair dice to show equal probabilities across outcomes.

Generate binomial random variates by simulating independent trials, compute the probability of exactly one six in ten dice rolls, and less than three sixes, then visualize with bar plots.

Generate samples from the Poisson distribution, visualize with histograms and bar plots, compare density across different lambda values, and use the probability quintile function to invert the Poisson cumulative distribution.

Learn to sample from a chi-squared distribution with R, using sums of squared standard normals, and compute density, distribution, quantiles, and the 95th percentile across degrees of freedom.

Understand the student's t distribution and its use in estimating the population mean when the standard deviation is unknown, and compare it with the normal distribution using R demonstrations.

Simulate the stochastic process by modeling stock trading with random samples, Brownian motion, and price changes; visualize returns and explore performance analytics.

Learn to compute and interpret confidence intervals in R, covering descriptive and inferential statistics, z-based and bootstrap methods, with normal and Poisson examples.

Learn to perform z-tests in R by comparing a sample to a population with known standard deviation, compute z-scores and one-tailed and two-tailed p-values, and implement a z-test function.

Learn to perform one-sample and two-sample t tests, decide between known or unknown standard deviation (Student's t with n<30), and interpret means using box plots with weight data.

Apply nonparametric Kolmogorov-Smirnov tests to compare a sample with a reference distribution or to compare two samples, using the empirical cumulative distribution function and KS test from the SAT package.

Learn how to perform Pearsons chi squared tests on a contingency table to assess independence between gender and smoking habit, with mosaic plots.

Learn to perform Wilcoxon rank sum and signed rank tests using the step package, analyzing Facebook likes, testing a median of 30, and comparing two fan pages.

Transform a purchase order dataset into transactions with a rules package; explore item associations and frequent sequential patterns, and learn to convert data frames and matrices into transactions in R.

Explore how to plot transactions and associations using the rules package in R. Display and summarize transactions, filter by size, and visualize item frequency with inspect, image, and itemFrequency.

Learn to mine associations in transactional data using the Apriori algorithm, discovering frequent item sets and strong rules with support, confidence, and lift.

identify and prune redundant association rules by comparing support and confidence thresholds, using superset and subset checks to reveal meaningful rules and remove duplicates.

Visualize association rules using the rules package, generate rules in a variable named rules (not pruned), and explore them with scatterplot, graph, and interactive visualizations.

Explore how the eclat algorithm generates frequent item sets by using a vertical data layout and intersecting transaction ID lists, then derives association rules from these sets.

Create a time series object in R with the ts function, load Taiwan Semiconductor data from csv, inspect its structure and class, and convert to an xts object.

Plot time series data using the plot.ts function, visualize trends and seasonal composition clearly, and compare multiple time series either in separate subfigures or a single figure with colors.

Select an Arema model to build a smoothing time series using exponential smoothing and account for autocorrelation, then use acf and pacf to choose p and q.

Select optimal p, d, q parameters and build an arima model from a time series using forecast and stats packages, then assess training accuracy and review the model summary.

Forecast future values with the forecast package in R, generate predictions from Arema/rhema model, summarize results, plot a line chart, and diagnose residuals with acf, box test, and tsdiag.

Predict stock prices using an Arema (arima) model by loading historical Facebook prices, fitting the model, forecasting future values, and visualizing results with plots and a returns histogram.

Scrape text from web pages and process it for analysis by reading text files and html content, extracting paragraphs with css selectors or xpath, and cleaning citations and spaces.

Learn to work with text corpora in the tm package, clean and stem text, build a term-document matrix with tf-idf weighting, and generate word clouds through code.

Explore cosine similarity to group documents by projecting them into a vector space, and apply latent semantic analysis using singular value decomposition to cluster themes.

Extract topics from a document corpus with latent Dirichlet allocation, modeling each document as a mix of topics and each topic by keywords, using a document-term matrix and Gibbs sampling.

Classify sentiment and mood in color transcripts using tidytext and Syuzhet, performing token extraction, stop-word removal, and per-call sentiment averages with AFINN and NRC lexicons.

Train multiple text classification models with the RTextTools package to categorize documents via a document matrix, using the CNAE-9 data set, and compare misclassification rates to identify the best method.