SynthesizePattern: Randomize a Time Series Pattern

In joshmarsh/TSTestDataUtil: TimeSeries Test Data Creation Utility

Description

Takes an xts object and performs time scaling to change the length of the whole pattern, time warping to randomly change the time positions of individual points to a limited extent, and noise adding to alter the amplitude of individual points randomly.

Usage

SynthesizePattern(pattern, scaling.factor = 4, warping.factor = "l",
  noise.factor = "l")

Arguments

pattern

An xts time series

References

Zhe Zhang, Jian Jiang, Xiaoyan Liu, Ricky Lau, Huaiqing Wang, and Rui Zhang. A real time hybrid pattern matching scheme for stock time series. In Proceedings of the Twenty-First Australasian Conference on Database Technologies - Volume 104, ADC <e2><80><99>10, pages 161<e2><80><93>170, Darlinghurst, Australia, Australia, 2010. Australian Computer Society, Inc.

Xueyuan Gong, Yain-Whar Si, Simon Fong, and Robert P. Biuk-Aghai. Financial time series pattern matching with extended ucr suite and support vector machine. Expert Syst. Appl., 55(C):284<e2><80><93>296, August 2016.

joshmarsh/TSTestDataUtil documentation built on May 19, 2019, 8:54 p.m.