Title
An Efficient Generalized Least Squares Algorithm for Periodic Time Series
Document Type
Student Presentation
Presentation Date
4-21-2014
Faculty Sponsor
Jaechoul Lee
Abstract
Over the course of a year time series data like daily temperature and sales fluctuate from highs to lows. When looking at the data values, one can see that they often form a periodic time series. The Generalized Least Square (GLS) estimators are used in many practices to estimate those fluctuations in the time series. However, calculating the GLS estimators can be computationally demanding and take a long time for big data. The point of this project is to develop a more efficient method to calculating the GLS estimators while still getting the same result. We plan on doing this by taking a big set of data and reformulating them into a very small data set so that it is easier to implement GLS computations.
Recommended Citation
Lee, Jaechoul; Dini, Anthony; and Negri, William, "An Efficient Generalized Least Squares Algorithm for Periodic Time Series" (2014). College of Engineering Presentations. 27.
https://scholarworks.boisestate.edu/eng_14/27