Regarding time series dimensionality reduction

venkateshkamini · Contributor Joined: 07 Jun 2011 Posts: 6

Dear all,

I am new to this forum but I think that this forum is very useful for people like me(beginner)

My question is about time series dimension reduction for representation. I think time series is having single dimension (ex:- gold price is having a single dimension ). Then how can we reduce dimension of it.

I read that reducing the number of data points is considered as dimension reduction.Is it right way to do that.

I would like to ask experts to give a brief explanation of it.

Thanks in advance.

editor · Posted: Sat Jun 11, 2011 1:55 pm Post subject: dimensionality reduction

How about using methods like SVD for dimensionality reduction?
http://en.wikipedia.org/wiki/Singular_value_decomposition

See also Keogh paper
Dimensionality Reduction for Fast Similarity Search in Large Time Series Databases
citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.36.4621.pdf