Turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

- JMP User Community
- :
- Discussions
- :
- Discussions
- :
- Re: Time Series, calculation lag

Topic Options

- Start Article
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Oct 26, 2013 12:46 PM
(4351 views)

Hello, I want to check if some temperature data that I have are correlated, and I suspect there is a time-dependent lag.

I have tried the time-series analysis using cross correlation. As a test of what the analysis does I entered the following data columns. There is full correlation for a lag of 7 rows:

A A+7

1 8

2 9

3 10

4 11

etc.

What I got just doesn't make sense. See the picture below. There should be some way to see a full correlation of the data for a lag of 7.

1 ACCEPTED SOLUTION

Accepted Solutions

Highlighted
Solution

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Hmm - I suppose this have to do with the nature of the data. When I try with random data and do a manual calculation (shifting numbers and calculating the integral of the product for each value) it seems to give me the same as JMP does. The lag is -4, which peaks as it should. See the picture.

The thing is that the time series should be filtered for any autocorrelation, something I've read is called whitening. Not mentioned and not possible in JMP in an automatic way as I see it. Anyway - JMP seems to catch my lag of -4 when the data is random.

Can anyone confirm this?

1 REPLY 1

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Hmm - I suppose this have to do with the nature of the data. When I try with random data and do a manual calculation (shifting numbers and calculating the integral of the product for each value) it seems to give me the same as JMP does. The lag is -4, which peaks as it should. See the picture.

The thing is that the time series should be filtered for any autocorrelation, something I've read is called whitening. Not mentioned and not possible in JMP in an automatic way as I see it. Anyway - JMP seems to catch my lag of -4 when the data is random.

Can anyone confirm this?