Denison Scientific Association: How to Analyze Time Series Data

We live in a world awash with data, on almost any topic imaginable. Increasingly, the answers scientists seek are contained in time series data, i.e., data sampled at different points in time. For such data, traditional statistical tools such as simple linear regression and ANOVA are insufficient, because data points can be dependent on other data points from the past. A more sophisticated model is required. In this talk, I’ll explain suitable models and will illustrate them with vignettes from my research, on problems related to policing, protests, and opioid overdose in Ohio. I will also summarize the content of Math 422: Time Series Analysis. This talk has no prerequisites, and students are welcome.