|
Activity Number:
|
365
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 2, 2016 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Business and Economic Statistics Section
|
|
Abstract #321355
|
View Presentation
|
|
Title:
|
Modeling Durations in High-Frequency Data Using Estimating Functions
|
|
Author(s):
|
Yaohua Zhang* and Jian Zou and Nalini Ravishanker and Aerambamoorthy Thavaneswaran
|
|
Companies:
|
University of Connecticut and Worcester Polytechnic Institute and University of Connecticut and University of Manitoba
|
|
Keywords:
|
Log ACD Duration Models ;
Estimating Equations ;
Fast Recursion ;
Time Series
|
|
Abstract:
|
Accurate modeling of patterns in inter-event durations is of interest in several applications and is important for capturing valuable information that facilitates decision-making. Since Engle and Russell (1998) first discussed autoregressive conditional duration models, several classes of models have been discussed in the literature for modeling durations (Thavaneswaran et al. 2014). Developing fast and accurate methods for statistical model fitting of long duration series is an interesting ongoing research problem. This talk describes an optimal semi-parametric modeling approach using martingale estimating functions, which only requires assumptions on the first few conditional moments of the durations. Three different methods for parameter estimation are illustrated and compared for simulated data and for real financial data. This talk also describes model selection and prediction for durations modeling.
|
Authors who are presenting talks have a * after their name.
