Feature Engineering with Sliding Windows and Lagged Inputs by Bryan Shalloway
Details
For some problems you may want to take a traditional regression or classification based approach while still accounting for the time-sensitive components of your data. In this post I will use the tidymodels suite of packages to:
- build lag based and non-lag based features
- set-up appropriate time series cross-validation analysis and assessment windows
- evaluate performance of linear regression and random forest models on a regression problem
***
Bryan has been a Data Scientist at NetApp for three years and worked on problems primarily in customer support and sales operations (e.g. support ticket prioritization, sales forecasting, pricing analysis). He holds a Master's from NC State's Institute for Advanced Analytics and a Bachelor's in Cognitive Neuroscience from Washington University in St. Louis. Prior to his work in data science, Bryan was an Education Analyst in Colorado as well as a high school math teacher with Durham Public Schools.
You can connect with Bryan on Twitter @brshallo or by sending him a note on LinkedIn. You can also read his blog at https://www.bryanshalloway.com/