|
Title:
|
INTRADAY FUTURES PRICE SERIES SMOOTHING |
|
Author(s):
|
idrina Pabarkait?, Rimvydas Simutis, Borisas Bursteinas, Aistis Raudys |
|
ISBN:
|
978-989-8704-10-8 |
|
Editors:
|
Ajith P. Abraham, Antonio Palma dos Reis and Jörg Roth |
|
Year:
|
2014 |
|
Edition:
|
Single |
|
Keywords:
|
Moving average, smoothness, accuracy, weight optimization. |
|
Type:
|
Full Paper |
|
First Page:
|
83 |
|
Last Page:
|
90 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
We analyse different methods for futures intraday time series smoothing. In trading community these methods are named moving averages, in statistics autoregressive models and in signal processing - filters. Smoothness and accuracy are two important criteria for quality evaluation. Smoothness is an important characteristic that can reduce trading costs while accuracy can help to detect trend earlier and profit more. Unfortunately, research in this area focuses on accuracy or smoothness separately, but we look at this as a multi criteria problem. In this study we define several smoothness levels and try to find the most accurate method for each smoothness level. This way we can compare various smoothing techniques like-to-like. We also propose a new smoothing method with optimised coefficients. We use 60 futures traded in US and EU exchanges and 1000 randomly generated datasets. We examine methods performance on intraday frequencies ranging from 1 minute to daily data. New method outperforms other methods 99% on the real world out of sample data. |
|
|
|
|
|
|
