ModifiedDTW 1.1.1

ModifiedDTW

Description

This is a tutorial project for finding optimum lag using the Modified Dynamic time warping along with the links.

Requirements

Use the package manager pip to install all the requirements. Just follow the commands below:

Install dtaidistance:

pip install dtaidistance

Cost matrix based on dtaidistance (https://pypi.org/project/dtaidistance/)

Install pandas:

pip install pandas

Data requirements:

Both timeseries should be in pandas series format with same datetime index

Usage

from ModifiedDTW.ModifedDTW import MDTWlag

series1 = Droughtindex1  # Cause
series2 = Droughtindex2  # Effect

Results = MDTWlag(series1, series2, Maximumwindowsize, LagRange)

Distanceatdifferentlags = Results[0]
OptimumLag = Results[1]
OptimumLinks = Results[2]

MaximumWindowSize defines the window within which the DTW path operates (default = 12).

LagRange specifies the maximum lag up to which the optimum lag will be searched (depends on the system under study; Default use MaximumWindowSize = LagRange).

Distanceatdifferentlags outputs a dataframe with total path distance at different lag.

OptimumLag is the propagation lag based on minimum path distance.

OptimumLinks is the datetime of links between the two indices at optimum lag.

Note: This package uses forward only band (Bilal-Gupta Band) preferred for analysis of links/lag between meteorological to agricultural/hydrological indices.

Cite

Bilal, S. B., & Gupta, V. (2026). A novel multi-link approach to drought propagation analysis using modified dynamic time warping. Journal of Hydrology, 135604. (https://doi.org/10.1016/j.jhydrol.2026.135604)

Contributing

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.