[Traminer-users] Formal definition of the Chi-Square distance metric?

[Gilbert Ritschard]

Hi Laurence,

The Chi-squared measure implemented in TraMineR is described in the JRSSA article
http://dx.doi.org/10.1111/rssa.12125

and you find more details (e.g. formula) in the LIVES Working Paper
https://www.lives-nccr.ch/sites/default/files/pdf/publication/33_lives_wp_studer_sequencedissmeasures.pdf

By the way, I strongly encourage you to ask your question on stackexchange (see http://traminer.unige.ch/contrib.shtml ) where you will reach more people.

All the best.
Gilbert

From: Traminer-users <traminer-users-bounces at lists.r-forge.r-project.org> On Behalf Of Laurence Droy
Sent: lundi 29 octobre 2018 22:14
To: traminer-users at lists.r-forge.r-project.org
Subject: [Traminer-users] Formal definition of the Chi-Square distance metric?

Dear All, 

I'd like to use the TraMiner package's Chi-square distance measure of the dissimilarity between sequences. 

However, I haven't been able to find a formal definition of how this is actually implemented and I've found the source code hard to interpret. 

Does anyone know the definition of the measure as used in Traminer? 

I'm hoping for something of the form:

D(i,j) = function-definition

Where i and j are a pair of sequences. 

Id be very grateful if someone could help me with this. 

Many thanks, 

Laurence Droy