Predictive Analytics With R

Learn how predictive modeling uses data mining, machine learning, and statistics to analyze historical data and predict future outcomes by identifying independent and dependent variables and their correlations.

Discover why predictive analytics matter for marketers, turning big data into actionable insights with statistics, data mining, and machine learning to forecast future customer behavior from past actions.

Explore how predictive analytics improves fraud detection, credit risk, and identity verification across businesses. Discover applications in direct marketing, customer retention, clinical decision support system, and movie recommendations.

R is a free open source language for data analysis, enabling analytics, insights, graphs, and simulations; implemented in C and extensible with functions and code in C, C++, or Fortran.

Explore the RStudio interface, including the console, syntax-highlighted editor, script and history panes, environment, packages, and the plot area, and learn how to run and save code.

Discover why R enables fast, accessible predictive modelling by bridging ease and sophistication, offering quick data access, simple sharing, and ready-to-use predictive analytics for decision makers.

Learn the mean as the average of observations, shown with its formula and an R studio example using a data vector to compute the mean function, yielding 5.

Identify the mode by examining frequency counts in a small dataset; 13 appears most frequently (four times) among 13, 14, 16, 18, 21.

Discover how to compute the median by sorting observations and selecting the middle value, or averaging the two central values for even datasets, using the median function in R.

Understand variance as the spread of data around the mean. Use R's var function on a sample dataset to compute it, with a mean around four and variance about 6.6.

Explore standard deviation as a measure of dispersion in a dataset, relate it to variance, and show how the square root of variance yields the standard deviation in R.

Explore skewness in data distributions, identifying positive and negative skewness, symmetry around the mean, and how to compute skewness in R with a data vector and the skewness function.

Explore the kurtosis concept and how distribution shapes indicate positive, neutral, or negative kurtosis, with hands-on RStudio implementation using a predefined function to calculate and visualize the distribution.

Explore how correlation measures the relationship between two or more variables, interpret positive, negative, and zero correlations, and compute the correlation coefficient with a built-in function.

Master group manipulation in predictive analytics with R by learning to split data into sections by metrics, apply transformations to each group, and recombine using aggregate, filter, and select functions.

Explore group manipulation in predictive analytics with r using lapply and apply to perform data-wide calculations on matrices and lists, including computing means.

Use the aggregate function to compute group-based statistics on categorical data, demonstrated with the diamonds dataset to compare mean price by cut, and horsepower by cylinder with na.rm = TRUE.

Master data manipulation for predictive analytics with R using the dplyr package to filter flights, select columns, mutate duration, and chain together arrange operations.

Explore string manipulation concepts in R, learning how to build and decompose strings, use regular expressions on text data, and apply key string functions.

Learn practical string manipulation in R through hands-on demonstrations of grep, grepl, gsub, tolower, toupper, substr, and strsplit for pattern matching, editing, and splitting data.

Learn to transform data in R by modifying an existing column with the transform function, and split data by a column such as age or gender, without adding new columns.

Explore how missing values in a small R dataset affect computations, and learn to identify, remove, or impute them with zero, mean, or column maximum.

Learn the concept of hypothesis testing by comparing sample data to population behavior, define null and alternate hypotheses, and use significance level and p-values to draw conclusions.

Explore hypothesis testing with mean and standard deviation, p-values at 5% significance, and decisions to reject or fail to reject the null hypothesis using a distribution function.

Explore the chi-square test concept as a goodness-of-fit test for categorical data, comparing observed to expected frequencies under a null hypothesis; compute the chi-square as sum of squared standardized differences.

Apply the chi-square goodness-of-fit test in R to compare observed eye-color frequencies to population proportions, using marginal and proportional tables to evaluate distribution.

Explore the t-test concept for comparing the means of two groups, focusing on small samples (n<30) to set up hypothesis tests about their means.

Apply a two-sample t-test in R to compare sleep outcomes between groups using the sleep dataset, and interpret p-values and confidence intervals to assess significance.

Explore the z-test concept for hypothesis testing with known standard deviation and large samples, using the standard normal distribution to accept or reject the null at 0.05 percent.

Perform a z-test in R to assess medication effects on intelligence, defining hypotheses, setting a 5% significance level, computing the z-score and p-value, and interpreting the 95% confidence interval.

Explore how regression analysis models the relationship between two variables, identifying dependent and independent variables, and compare linear, multiple, and logistic regression while assessing statistical significance of estimated relationships.

Learn simple linear regression to analyze the association between numeric independent variable and a dependent variable. Interpret the regression equation, including the intercept and slope, for predictive analytics in R.

Apply linear regression in R on the trees dataset to predict height from girth, using lm and summary outputs to interpret intercept, slope, residuals, R-squared, and confidence intervals.

Learn the concept of multiple regression, a powerful technique that uses several independent variables to predict a single dependent outcome, with an intercept and flexible variable counts.

Implement multiple regression in R to predict a dependent variable (retention) using 11 predictors, interpret coefficients and significance, and apply backward and forward stepwise selection to identify key predictors.

Explore logistic regression for binary outcomes by modeling odds of events, using binary, ordinal, or multinomial outcomes, with categorical or continuous predictors, illustrated by credit default and direct mail.

Learn to implement logistic regression in R with a graduate admissions example using GPA and prestige, including data prep, fitting a binomial glm, interpreting coefficients, and generating predictive probabilities.