Data 360

Data 360 Certificate Program

Designed to enhance functional literacy in critical business analytics, increase the accuracy of predictions and make better, more agile business decisions. Course work included advanced techniques in prescriptive analytics including optimization and modeling as well as understanding the interdependent effects of multiple decisions.

We learned, or in my case re-acquainted, ourselves with scientific methods for data analysis and visualization and gained a more complete understanding of risk and probability, using statistical models to optimize outcomes for complex—and often simultaneous—business decisions.

---

Syllabus

Understanding & Visualizing Data (SHA571):

This course was designed to provide a basic grounding in many statistics concepts and get students to move beyond making decisions focused solely on averages and develop a working familiarity with the grounding principles of data analysis. We learned to derive the greatest benefit possible from the data available while ensuring that the conclusions drawn remain valid. Through exercises and projects we applied a decision-making framework where we interacted with the data to achieve the best outcome.

Lessons covered how to:

·      Identify and collect data that can be used to address any given business problem

·      Recognize and mitigate potential bias when generating a data sample

·      Develop statistical summaries and data visualizations to understand how variables impact outcomes

·      Evaluate decisions by looking at key performance measures and determining their implications for stakeholders

Implementing Scientific Decision Making (SHA 572):

Building upon the first course, we moved on to how to examine sample data scientifically so to limit any generalizations to only the patterns that have the strongest statistical support.

Lessons covered how to:

·      Formulate a question as a null and alternate hypothesis

·      Calculate a test statistic from sample data

·      Identify the statistical test most appropriate for testing your hypothesis

·      Determine the likelihood of finding a result at least as extreme as the test statistic assuming the null hypothesis

Using Predictive Data Analysis (SHA 573):

The sheer variety of sources and types of data that can aid in decision making are almost overwhelming. The key to making good use of the data lies in knowing what specifically to pay attention to, understanding the relationships and variables among the data, and making the right connections.

Experience is essential to knowing and making educated guesses about what to pay attention to. The focus of this course was to gain familiarity with statistical methods that can provide an advantage over relying on gut instinct alone.

In this course we learned to identify uncertainty in a business decision, and to choose variables that help reduce uncertainty. By the end of this course, we developed a robust decision model that could be used to make predictions related to decisions. Along the way, we clarified and enhanced our understanding of the factors that influence possible outcomes from the decision.

Lessons covered how to:

·      Determine the degree of uncertainty in your decision and determine the impact of this uncertainty

·      Identify data relationships to reduce uncertainty

·      Create a regression model that looks at attributes of variables driving the decision

·      Refine your regression model to improve its validity

·      Create a convincing argument for the validity of your model

·      Make a prediction or an estimate using your model

Modeling Uncertainty & Risk (SHA 574):

This course required us to the use foundations in probability to describe risk mathematically and incorporate those calculations into our decisions so you can take them to the next level. We worked through increasingly complex modeling situations, learning to use estimates of probable future outcomes for Go/No-Go decisions and how to run a Monte Carlo simulation. The result was the capacity to examine outcomes that vary based on multiple, interdependent decisions.

Lessons covered how to:

·      Calculate marginal value for a binary decision

·      Determine optimal values for a repeating, sequential decision

·      Build risk aversion into your model

·      Calculate utility for a given decision

·      Develop and use a Monte Carlo simulation

·      Perform sensitivity analysis

·      Use expected utility to accommodate risk

Optimization & Modeling Simultaneous Decision Making (SHA 575):

The final and most complex course tackled the reality that we don’t often have the luxury of making one decision at a time but instead face multiple decisions at once, in highly complex situations where each decision has potentially far-reaching impacts. The aim of this course was to provide us with a robust, quantifiable understanding of these ripple effects in order to meet business objectives and raise the odds of decision-making success. Through exercises and projects we learned to create and use data models for optimizing decision making in situations where resources are constrained—and two or more decisions whose consequences interact must be made simultaneously.

Lessons covered how to:

·      Create an optimization model

·      Solve a linear model in Excel using the Solver add-in

·      Make tactical decisions using shadow prices

·      Set prices in a non-linear model

·      Approximate an optimal solution for a non-linear model

·      Use summary statistics to approximate optimal results for a stochastic problem