Classification and Planned Experiments

Overview

Welcome to Classification and Planned Experiments. This course will first contrast regression models with classification models, which have broad application in machine learning. It will then introduce basic classification techniques, focusing on K-nearest neighbor, and logistic regression. You will examine data visualizations and see how setting hyperparameters or estimating parameters supports interpretation and effective classification. The course will then address another powerful field of applied statistics called experimental design, which is concerned with running controlled tests (experiments) to try to understand causal relationships between factors of interest. Several types of designs will be introduced, including ones that use computer modeling. You will learn the principles of experimental design and work through several examples to help you understand how to actually set up, run and analyze these experiments leveraging data.

Syllabus

Course Introduction

This Specialization covers the use of statistical methods in today's business, industrial, and social environments, including several new methods and applications. H.G. Wells foresaw an era when the understanding of basic statistics would be as important for citizenship as the ability to read and write. Modern Statistics for Data-Driven Decision-Making teaches the basics of working with and interpreting data, skills necessary to succeed in Wells’s “new great complex world” that we now inhabit. In this course, learners will gain an ability to execute basic classification techniques, including the use of R and Python; apply the principles of experimental design; and demonstrate usage of propensity scores, causal inference, and counterfactuals in causal learning.Learn more about the instructors who developed this course. Read the instructor bios and review the learning outcomes for the course.

Introduction to Planned Experiments

This module will focus on experiment design, fraction factorial design, and computer experiments. We will review a brief history of experiment design, and relevant terminology. We will review guidelines for conducting and analyzing experiments and applying design to computer models.