Mathematics Colloquium

Add to Calendar 03/16/2017 15:35:0003/16/2017 16:35:0015Mathematics ColloquiumSpeaker: Seth Gannon Advisor: Dr. Maher Qumsiyeh Title: Bootstrapping General ARIMA Models Abstract:  The Bootstrap of Efron (1979) has been shown to be an effective method for estimation and testing purposes. In regression models and in autoregressive time series models we resample the residuals to create new observation and use the ARIMA model to get estimates for parameters. This will not work in a moving average or mixed ARIMA models because the residuals are correlated. Due to this a different bootstrap approach must be used to deal with the dependency, known as, the non-overlapping block bootstrap method. In this paper we will show how the non-overlapping block bootstrap method can be used for parameter estimation and for forecasting in a moving average and in a mixed autoregressive-moving average models for simulated data. The non-overlapping block bootstrap method will be compared with the Box-Jenkins methodology for parameter estimation and forecasting. We will also compare the length of the confidence intervals for the parameters and forecasted values using the traditional methods and the non-overlapping block bootstrap method. All programming was done using the statistical software package (SAS). Refreshments are available at 3:00 PM in SC 313F. The department colloquia are held every Thursday (excluding holidays) at 3:35 pm in room SC 323 unless otherwise noted. All are invited to attend. Science Center Room 323Paul Eloepeloe1@udayton.eduNo03/16/2017

Thursday, March 16

Time: 3:35 p.m. — 4:35 p.m.

Location: Science Center Room 323

Tags:  Arts and Sciences, Colloquia, Mathematics, Sciences

Cost:  Free

Speaker: Seth Gannon

Advisor: Dr. Maher Qumsiyeh

Title: Bootstrapping General ARIMA Models

Abstract: 

The Bootstrap of Efron (1979) has been shown to be an effective method for estimation and testing purposes. In regression models and in autoregressive time series models we resample the residuals to create new observation and use the ARIMA model to get estimates for parameters. This will not work in a moving average or mixed ARIMA models because the residuals are correlated. Due to this a different bootstrap approach must be used to deal with the dependency, known as, the non-overlapping block bootstrap method.

In this paper we will show how the non-overlapping block bootstrap method can be used for parameter estimation and for forecasting in a moving average and in a mixed autoregressive-moving average models for simulated data. The non-overlapping block bootstrap method will be compared with the Box-Jenkins methodology for parameter estimation and forecasting. We will also compare the length of the confidence intervals for the parameters and forecasted values using the traditional methods and the non-overlapping block bootstrap method. All programming was done using the statistical software package (SAS).

Refreshments are available at 3:00 PM in SC 313F.

The department colloquia are held every Thursday (excluding holidays) at 3:35 pm in room SC 323 unless otherwise noted. All are invited to attend. 

Contact Information:

Name:  Paul Eloe
Email:  peloe1@udayton.edu