|
Activity Number:
|
574
- Recent Advances in Software
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, July 31, 2019 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistical Computing
|
|
Abstract #304489
|
|
|
Title:
|
Analytical Likelihood Derivatives for State Space Forecasting Models
|
|
Author(s):
|
Jonathan Hosking* and Ramesh Natarajan
|
|
Companies:
|
Amazon.com and Amazon.com
|
|
Keywords:
|
time series;
maximum likelihood;
computation
|
|
Abstract:
|
State space models are a flexible and widely used family of statistical models for time series analysis and forecasting. Calibration of the models to historical data is greatly facilitated by the availability of analytical derivatives of the log-likelihood function. We ave obtained a new expression for these derivatives in terms of quantities routinely computed in Kalman filtering and smoothing. This result makes it straightforward to construct an optimization method based on gradient descent using analytical log-likelihood derivatives. We present the derivation and give some examples of the gain in speed of calibration when analytical derivatives are used.
|
Authors who are presenting talks have a * after their name.
