Presented By
O’Reilly + Cloudera
Make Data Work
March 25-28, 2019
San Francisco, CA

Time series forecasting using statistical and machine learning models: When and how

Ying Yau (Walmart Labs)
4:20pm5:00pm Wednesday, March 27, 2019
Average rating: ***..
(3.29, 7 ratings)

Who is this presentation for?

  • Data scientists, statisticians, VPs, directors, and managers of data science, and anyone who needs to conduct time series forecasting



Prerequisite knowledge

  • A working knowledge of Python
  • A basic understanding of classical linear regression models, machine learning techniques (SVM, tree-based models, and neural network based models), and time series forecasting

What you'll learn

  • Explore the mathematical formulation and coding implementation of autoregressive integrated moving average (ARIMA) models, vector autoregressive (VAR) models, and neural network-based models
  • Understand the pros and cons of the application of ARIMA, VAR, regime-switching models, random forest, and neural network-based models to time series forecasting


Time series data is ubiquitous: daily term structure of interest rates, tick-level stock prices, daily foreign currency exchange rates, weekly initial unemployment claim, monthly company sales, daily foot traffic recorded by mobile devices, and daily number of steps taken recorded by a wearable, just to name a few. Some of the most important and commonly used data science techniques in time series forecasting are those developed within the fields of statistics and machine learning. A few basic time series statistical and machine learning modeling techniques for forecasting should be included in any data scientist’s toolkit.

Jeffrey Yau discusses the applications of statistical time series models, such as ARIMA, VAR, and regime-switching models, and machine learning models, such as random forest and neural network-based models, to forecasting problems. You’ll learn the advantages and disadvantages when using each of these models in time series forecasting scenarios and explore real-world applications, demonstrated using Jupyter notebooks, that illustrate these techniques. And along the way, Jeffrey touches on exploratory time series data analysis.

Photo of Ying Yau

Ying Yau

Walmart Labs

Jeffrey Yau is a distinguished data scientist at WalmartLabs, where he leads data science for the store technology department. Previously, he was the chief data scientist at AllianceBernstein, a global asset-management firm; the vice president and head of data science at Silicon Valley Data Science, where he led a team of PhD data scientists helping companies transform their businesses using advanced data science techniques and emerging technology; the head of risk analytics at Charles Schwab; director of financial risk management consulting at KPMG; assistant director at Moody’s Analytics; and assistant professor of economics at Virginia Tech. Jeffrey’s active in the data science community and often speaks at data science conferences and local events. He has many years of experience in applying a wide range of econometric and machine learning techniques to create analytic solutions for financial institutions, businesses, and policy institutions. Jeffrey holds a PhD and an MA in economics from the University of Pennsylvania and a BS in mathematics and economics from UCLA.

Comments on this page are now closed.


03/29/2019 10:20am PDT

can you share your presentation slides.