Automated shifting correlation test

Click For Summary
SUMMARY

The discussion centers on performing automated shifting correlation tests between two datasets using R. The primary function mentioned for this purpose is the R function ccf, which calculates cross correlations in time series data. Participants suggest searching for resources on "cross correlation time series" for further guidance. The conversation also touches on the potential applications of this analysis in fields like finance.

PREREQUISITES
  • Understanding of time series analysis
  • Familiarity with R programming
  • Knowledge of the ccf function in R
  • Basic concepts of correlation and data visualization
NEXT STEPS
  • Research the ccf function in R for cross correlation analysis
  • Explore graphical representation techniques for time series data
  • Learn about lagged variables and their impact on correlation
  • Investigate advanced time series analysis methods in R, such as ARIMA models
USEFUL FOR

Data analysts, statisticians, and researchers interested in time series analysis and correlation testing, particularly those working with financial data or seeking to automate their analysis processes.

austinboston
Messages
2
Reaction score
0
Hello everyone I'm new here:

So I was about to write some SAS or R to do the following, but now I'm getting old enough to realize most things I think of were thought of before (usually by people way smarter than me). I've done some googling, but I haven't found the golden answer yet.

I'd like to test correlation between two data sets but I want to test a ton of "shifts" between the two sets. I want to automatically play with the lead and lag of the two sets and look at some sort of clever graphical representation of the relationships.

If anyone out there can point me to some R/SAS functions that is great, but even better would be the 10,000 foot guide of the basics for this type of question. The high level type guidance is usually what I have trouble googling up, instead I usually get too specialized information.

Thanks,
Austin
 
Physics news on Phys.org
austinboston said:
Hello everyone I'm new here:

So I was about to write some SAS or R to do the following, but now I'm getting old enough to realize most things I think of were thought of before (usually by people way smarter than me). I've done some googling, but I haven't found the golden answer yet.

I'd like to test correlation between two data sets but I want to test a ton of "shifts" between the two sets. I want to automatically play with the lead and lag of the two sets and look at some sort of clever graphical representation of the relationships.

If anyone out there can point me to some R/SAS functions that is great, but even better would be the 10,000 foot guide of the basics for this type of question. The high level type guidance is usually what I have trouble googling up, instead I usually get too specialized information.

Thanks,
Austin

Hi Austin,

Yeah, you're right, I think most of us have been in the situation "Oh! My God! I just invented the wheel!" :smile:

What you describe is called cross correlations in time series and the function for this in R is named ccf

About the guide, just search for 'cross correlation time series' and you will find tons.

PS: Just guessing, but If you intend you get rich with this in the stock market just save your money and thank me later :-p
 
viraltux said:
Hi Austin,

Yeah, you're right, I think most of us have been in the situation "Oh! My God! I just invented the wheel!" :smile:

What you describe is called cross correlations in time series and the function for this in R is named ccf

About the guide, just search for 'cross correlation time series' and you will find tons.

PS: Just guessing, but If you intend you get rich with this in the stock market just save your money and thank me later :-p

That was what I needed, thanks. I had some fun playing with sin & cos cross correlation in R to get my feet wet. Ha, this idea did come to me while in a conversation about the stock market, but I think I have a problem at work I can use this for as well. Thanks again.
 

Similar threads

  • · Replies 4 ·
Replies
4
Views
6K
  • · Replies 21 ·
Replies
21
Views
4K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 14 ·
Replies
14
Views
3K
  • · Replies 5 ·
Replies
5
Views
3K
  • · Replies 11 ·
Replies
11
Views
2K
  • · Replies 1 ·
Replies
1
Views
6K
  • · Replies 3 ·
Replies
3
Views
1K
  • · Replies 4 ·
Replies
4
Views
7K