Register for Training
Course size is limited to encourage interaction and close learning with instructor.
Breakfast, lunch, and coffee will be provided throughout.
For more information please contact:
Pete Ungberg
Product Manager
Phone: +1 212 986 9300
E-mail: [email protected]
R instructors
Daniel L. Jerrett, Ph.D.
Daniel teaches econometrics and forecasting at the University of Colorado Denver, within the School of Business, the school Public Health as well as the JP Morgan Commodities Center. He has over 15 years of econometric experience in both the private and public sector.
Course Descriptions
Day 1
Python Bootcamp
Monday, April 29th
8:30 AM – 5:00 PM
"Python Bootcamp" introduction to the power of Python language for data handling, manipulation, analysis and presentation. "Python Bootcamp" is highly recommended even for the most experienced user, as Haver database integration and an introduction to Python visualization will be covered in detail.
Topics will include:
- Intro to Python environment via Anaconda, and Jupyter notebooks
- Loading data into Python and spreadsheet import
- Data visualization and graphing featuring Matplotlib and Seaborn
- Haver database link with Python - database queries, arguments, and search functionality
- Time-series data analysis - summary stats, correlograms, tests for auto-correlation
- Data transformations - natural logs, working with differences
- Regressions in Python - OLS model and applied examples
Days 2 & 3
Applied Time Series Pt. 1
Tuesday, April 30th to Wednesday, May 1st
8:30 AM – 5:00 PM
"Applied Time Series with Python" is the follow up to "Python Bootcamp". The objective is to introduce fundamental time-series concepts and teach their implementation with a wide selection of packages available in Python. Real-world economic and financial data is used throughout the two days giving participants the opportunity to explore data and view results directly related to their line of work.
This two-day course is designed to be applied with a hands-on approach taken. Applicable time-series theory is introduced, but formal proofs are not explored in depth. The objective of the course is to leave participants with a catalog of time-series models and the knowledge to know when and how to apply each. Throughout the course, participants will be given additional resources and references for further exploration.
Day 2 topics will include:
- Univariate models in Python
- Model selection: R-squared, Mean Squared Error, Akaike and Schwarz information criteria
- Serial correlation
- Unit root tests
- Forecast evaluation
- Introduction to multivariate models
Day 3 topics will include:
- Vector auto-regressions (VAR) - multivariate mode
- Co-integration and Engle-Granger test
- Vector Error Correction Models (VECM)
- Principal component analysis and application to investments and forecasting
Required Equipment and Software
Participants are expected to bring their own laptops equipped with Python, Anaconda, Jupyter, Excel, and Adobe Acrobat Reader.* Other course materials will be provided.
* A limited number of laptops may be loaned on-site in case of technical issues
Cancellation Policy
Cancellations received prior to two full business days before the date of the seminar will be honored. For cancellations received after the two day deadline but before the date of the seminar, the full fee less $150 will be converted to a transferable, nonrefundable credit to be applied toward a future seminar. Any credits issued must be used within one year. If notification of cancellation is not received prior to the first day of the seminar the full fee is payable.
Note: Colleague substitution permitted with no penalty.
or call +1 212 986 9300 or email Pete Ungberg
