Pattern distance of time series
Gang Rong · Journal of Zhejiang University(Engineering Science) · 2004
The pattern model representation (PMR) of time series was proposed for measuring the trend similarity of the time series. PMR is based on piecewise linear representation and is effective at describing the tendency of time series. Because normalization was unnecessary, series pattern distance (SPD) was calculated rapidly. SPD overcomes the problem of time series mismatch based on point distance. According to the numbers of time series' segmentations, pattern distance has multi scale feature and can reflect different similarity of time series under various analyzing frequency.