Does the correlation of matrices have explanatory power when building a pattern recognition model?

I'm using 8 different variables (with daily observations) with the purpose to compare different months across the historical data. For that purpose I calculate the correlation between each month and the historical months in the data and then calculate the Euclidean distance in order to find the closer month.

Does it make sense? Is there any literature regarding such experiments?

• Would it be right to say that you are trying to find seasonal (periodical) patterns in returns? Sep 18 '14 at 14:18
• Right! I'm trying to find seasonal patterns on some macroeconomic variables, with the purpose to find the best asset allocation. (for instance in a simple way, which asset class is going to perform better next month: Equities or Bonds) Sep 18 '14 at 14:25

2 Answers

There is a vast literature on modelling time-series with periodcities. Rob Hyndman is one of the leading reseaerchers in this area. He has published the R package forecast and a free online text book on this subject (with another package and R code in the book). Your task is covered starting here.

Analyzing seasonal time series "by hand" is not a good idea because there is a lot of time series machinery developed just for that. A simple example in R can be found here.

You can apply clustering if it feels more natural but the main question is whether your model works out of sample.

• I'm using the K-nearest neighbour and in order to find the most similar patterns I'm using the Euclidean distance. Do you think I'm doing it on a wrong way? Thanks in advance James Sep 18 '14 at 14:53