Presented By O’Reilly and Cloudera
Make Data Work
September 11, 2018: Training & Tutorials
September 12–13, 2018: Keynotes & Sessions
New York, NY

Modeling time series in R

Jared Lander (Lander Analytics)
3:30pm–4:10pm Thursday, 09/13/2018
Data science and machine learning
 Location: 1A 06/07 Level: Beginner
Secondary topics:  Temporal data and time-series analytics
Average rating: *****
(5.00, 3 ratings)

Who is this presentation for?

  • Data scientists, analysts, machine learning practitioners, sysadmins, and DBAs

Prerequisite knowledge

  • A basic understanding of R and math

What you'll learn

  • Explore time series methods
  • Learn the practical coding skills of fitting time series models and the code to generate interactive plots

Description

Temporal data is being produced in ever-greater quantity by processes such as telemetry data, the internet of things, sensors, server workloads, and transactions, but fortunately our time series capabilities are keeping pace. Jared Lander explores techniques for modeling time series, from traditional methods such as ARMA to more modern tools such as Prophet and machine learning models like XGBoost and neural nets. Using bike share data as a use case for the various techniques, Jared covers just enough math for each of the models and details in depth the R code for training these models to data and generating forecasts. Along the way, Jared also examines R objects designed specifically for time series and makes use of interactive visualizations.

Topics include:

  • xts objects
  • Mapping
  • Interactive time series plots
  • Stationarity
  • Seasonality
  • Autocorrelation
  • Autoregressive integrated moving average models
  • Generalized additive models with Prophet
  • Neural networks
  • XGBoost
  • The forecast package
Photo of Jared Lander

Jared Lander

Lander Analytics

Jared P. Lander is chief data scientist of Lander Analytics, where he oversees the long-term direction of the company and researches the best strategy, models, and algorithms for modern data needs. He specializes in data management, multilevel models, machine learning, generalized linear models, data management, visualization, and statistical computing. In addition to his client-facing consulting and training, Jared is an adjunct professor of statistics at Columbia University and the organizer of the New York Open Statistical Programming Meetup and the New York R Conference. He is the author of R for Everyone, a book about R programming geared toward data scientists and nonstatisticians alike. Very active in the data community, Jared is a frequent speaker at conferences, universities, and meetups around the world and was a member of the 2014 Strata New York selection committee. His writings on statistics can be found at Jaredlander.com. He was recently featured in the Wall Street Journal for his work with the Minnesota Vikings during the 2015 NFL Draft. Jared holds a master’s degree in statistics from Columbia University and a bachelor’s degree in mathematics from Muhlenberg College.