Decompose Signal Component

@Maarit

@Corey

Hi there,

I have been trying to run your workflow "Accessing, Transforming and Modeling Time Series"

The “Decompose Signal” node is set to Max Lag 24 and it is not running:

“Can only compute partial correlations for lags up to 50% of the sample size. The requested nlags 24 must be < 24.”

Ok… Try with 23 since table is dim=48. Result is KO.

Only works for mlag<=18.

First Lag is set to First Lag = 12 after first “Inspect Seasonality” component.

Table resulting after “Removing Seasonality” metanode is not dim=48 anymore but 36 (48-First Lag)

I have gone deeper and found that within the “Remove Seasonality” metanode and then again within the “Remove Seasonality” component there is a “Missing Value” node that is not doing what it is tagged to do.

It is tagged “replace missing vals with 0” but it is removing rows with missing values.

That is why the input table to second “Inspect Seasonality” component is dim=36 and only works for Max Lags <= 18.

Am I right?

Hi @Iceman,

I’m sorry someone hasn’t responded to this sooner!

I checked the “Accessing , Transforming, and Modeling Time Series” workflow you mentioned and I believe you are correct. It looks as if the Missing Value node may have been improperly setup. When you ran the workflow again after editing the node to correctly replace the missing values with zero, were you able to configure your Max Lags to equal 24?

Cheers,
Dash

3 Likes

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.