[Bug]: Time Series Analysis Stationarity transformation

[wooyoung32]

JASP Version

0.18.3

Commit ID

No response

JASP Module

Time Series

What analysis are you seeing the problem on?

Transformation under stationarity

What OS are you seeing the problem on?

macOS Intel

Bug Description

I've noticed that when I perform data transformation, I get identical results using the methods "Detrend using linear regression + Log" and "Detrend using linear regression". Initially, I thought this might be due to the "Detrend using linear regression" step generating negative data values, causing the "Log" step not to work. To check if the problem was the negative value of the data, I ran using "Difference"(which also produces negative values) instead of "Detrend using linear regression". However, the "Difference + Log" and "Difference" produced different results.

Expected Behaviour

Results from "Detrend using linear regression + Log" and "Detrend using linear regression" should differ.

Steps to Reproduce

1. Do "Detrend using linear regression + Log" transformation
2. Do "Detrend using linear regression" transformation
3. Compare the results
   Pennsylvania monthly offences data_2013.csv

...

Log (if any)

No response

Final Checklist

• I have included a screenshot showcasing the issue, if possible.
• I have included a JASP file (zipped) or data file that causes the crash/bug, if applicable.
• I have accurately described the bug, and steps to reproduce it.

[tomtomme]

@sophieberkhout
Can you help out here?
I can confirm this even for current 0.19 beta:

[wooyoung32]

Hi @tomtomme
Just checking in on the bug status. Is it still being worked on, or has it been resolved? Thanks!

[EJWagenmakers]

@sophieberkhout what's the current status of this one?

[sophieberkhout]

I believe I have found the problem and will be able to fix it soon.

[tomtomme]

still valid for 0.19.2 beta flatpak

@sophieberkhout
any update on this one?

jasp file to test:
Pennsylvania.monthly.offences.data_2013x.jasp.zip

[tomtomme]

almost finished via jasp-stats/jaspTimeSeries#38