Statistics for Data Science

Familiarize yourself with the basics of modeling and understand what it means, learn how to define the concept of an objective function for evaluating model performance, and understand how to carry out unsupervised/supervised analysis and the bias/variance tradeoff with an introduction to different model types.

Understand the concepts of random variables and hypothesis testing. Gain exposure to different data distributions and methods for hypothesis testing.

Explore distance and density-based clustering methods for exploratory analysis—k-means, hierarchical clustering, and DBSCAN—and exploratory analysis by selecting the appropriate clustering method to expand your knowledge of data sets.

Discover dimension reduction and learn how to apply principal components analysis (PCA) as a method, including the fundamentals of PCA to comprehend how its results have significance in terms of the original data and the creation of meaningful features from exploratory analysis that will help you perform supervised modeling.

Examine Method of Moments and learn how to use it to determine the linear model parameters, understand the assumptions and restrictions of a linear model, and evaluate the estimates and suitability in a linear model.

Learn to perform variable transformations and include interaction terms to improve model quality, discover and address issues with multicollinearity, and incorporate features from exploratory analysis when building a linear model.

Delve into the concept of a classification model while learning about the intricacies of logistic regression, outputs, and link functions. Understand the extension of binary logistic regression to multinomial logistic regression.

Examine the process of modeling categorical independent variables; evaluate the outputs of ANOVA, both from a traditional linear model and for determining if group means values are significantly different; and perform multiple pairwise comparisons and analyze models with multiple categorical predictors.