Bootstrap approximation on the statistics for the slope parameter of a stationary first-order autoregressive model

• Main Authors: Aldover, Ailinette Margarita G., Organo, Archimedes
• Format: text
• Language: English
• Published: Animo Repository 1992
• Subjects: Bootstrap theory (Nuclear physics); Statistical hypothesis testing; Approximation theory; Mathematical statistics
• Online Access: https://animorepository.dlsu.edu.ph/etd_bachelors/15964
• Institution: De La Salle University

The aim of this paper is to apply and examine the bootstrap approximation of the T-Statistic often used for testing hypothesis on the slope parameter, B in the stationary first order autoregressive model,Y t = + a + By t-1 + Et.The results obtained are compared with the Student-t approximation. Bootstrap is accomplished by resampling from the data values, with replacement and then recomputing the original statistic. It is a computer based simulation method which is used in assessing repeated sampling properties of a statistical procedure.A computer program, using Turbo Pascal Version 5.0, is formulated to provide all necessary simulations, calculations and estimations.