1. Introduction to Decision Science
Welcome to this course on Demystifying Decision Science! My name is Howard Friedman.
2. Welcome
Akshay Swaminathan and I co-authored a book called Winning with Data Science. Akshay previously led data science work at Flatiron Health and Cerebral and is a Knight-Hennessy scholar and PD Soros Fellow at Stanford University School of Medicine. I teach at Columbia University and have decades of experience leading data projects in the private and public sectors. We are very excited to share with you our expertise on how to understand and apply Decision Science.
3. What is Decision Science?
Decision Science combines principles from mathematics, statistics, psychology, economics, and computer science to help individuals and organizations make better decisions. Simply put, decision science is a systematic and data-driven approach to solving problems and optimizing outcomes.
4. Informed decisions
Decision science is about making informed decisions. This involves a structured process that typically includes defining the problem, gathering data, analyzing the data, developing alternative solutions, evaluating those alternatives, selecting the best option, and finally, implementing and monitoring the chosen solution.
5. Why perfect rationality isn't realistic
Traditionally, decision-making models assumed people were perfectly rational. This meant they had complete information, could process it without bias, and chose the best option for what they were trying to optimize.
We now understand more clearly that humans operate with bounded rationality. We have limited cognitive resources, time constraints, and are influenced by emotions and biases. Decision science acknowledges these limitations and provides tools and frameworks to make better decisions despite these constraints
6. Data-driven decision-making
One of the core principles of decision science is the emphasis on data-driven decision-making. This means relying on data rather than intuition. By analyzing data, we can gain objective insights, identify trends, and make more accurate predictions. This approach is crucial since today we have access to vast amounts of information.
Imagine you are running a forecasting project at a finance company. You will be constantly asked to provide data-driven guidance on which investment will improve the company's growth and profitability while minimizing the exposure to risks. This is a perfect opportunity for Decision Science.
7. Unconscious bias
Even with the best data, our decisions can be influenced by unconscious biases. These are mental shortcuts that can lead to errors in judgment. Some common biases include confirmation bias in which we seek information that confirms our existing beliefs, anchoring bias where we over-rely on the first piece of information received, loss aversion where we avoid losses over gaining an equivalent amount, and availability heuristic where we overestimate the likelihood of events that are easily recalled. Decision science helps us to recognize and mitigate these biases.
8. Descriptive versus prescriptive
Finally, let's touch upon two key types of analytics used in decision science:
Descriptive analytics focuses on understanding the past. It uses techniques like data visualization and summary statistics to describe what has happened in the past.
Prescriptive analytics goes a step further by recommending future actions. It uses techniques like optimization and simulation to identify the best course of action. This is where the analysis of past data is used to project what is likely to happen in the future.
9. Decision Science as a framework
Decision science provides a powerful framework for making informed decisions. By understanding the decision-making process, acknowledging our limitations, and leveraging data-driven approaches, we can achieve better outcomes.
10. Let's practice!
Throughout this course, we'll explore various tools and techniques of decision science, empowering you to become more effective problem-solvers and decision-makers.