Cross section,Time Series and Panel data (balance&unbalance)

Discuss the jamovi platform, possible improvements, etc.
Post Reply
kalemguy
Posts: 3
Joined: Fri Oct 07, 2022 2:43 am

Cross section,Time Series and Panel data (balance&unbalance)

Post by kalemguy »

Hi,
I am a newbie in research...and my data will be unbalanced panel data.
Is Jamovi able to differentiate between Panel Data, Cross Sectional and Time Series? What is the default in Jamovi?
If it is able to do so, how can we do it? Is there a catch for it? For example, what is the sequence of the data (code, year, etc) that we must prepare prior to importing into Jamovi
As for Panel Data, how can we tell Jamovi that what is the "individual" and "time" (the cross-section or the time series)
Lastly, does unbalanced panel data can be executed by Jamovi? Or what should be done prior to importing unbalanced panel data into Jamovi if applicable?

Your Help is much appreciated
User avatar
jonathon
Posts: 2613
Joined: Fri Jan 27, 2017 10:04 am

Re: Cross section,Time Series and Panel data (balance&unbala

Post by jonathon »

hi,

i don't completely understand your questions, but jamovi doesn't support time series at this time.

cheers

jonathon
kalemguy
Posts: 3
Joined: Fri Oct 07, 2022 2:43 am

Re: Cross section,Time Series and Panel data (balance&unbala

Post by kalemguy »

Thank you for your reply,

I apologize, that I am not understandable,

Basically, I want to know how Jamovi can identify the dataset is cross-section or panel data?

Because according to the information I received, to analyse panel data in a statistical software such as STATA or R, we need to set our data into panel data. To set our data into panel data, we need to justify the cross-section and time dimension columns or variables.

For example, to set panel data in R, they use the command "pdata.frame(Gasoline, index = c("country", "year")" where Gasoline is the name of dataset, country is the identification for cross-section dimension and "year" is the identification for time.

Therefore, I would like to know whether we need to set up our data into panel data prior to analysis or not? If we are not require setting up our data into panel, how can Jamovi Identify that our data is Panel? Is there any sequence of column or variable that we need to follow?, for example the first column always cross-section dimension data and the second column is time dimension data

Please help me with the above matter
Thank you
User avatar
jonathon
Posts: 2613
Joined: Fri Jan 27, 2017 10:04 am

Re: Cross section,Time Series and Panel data (balance&unbala

Post by jonathon »

hi,

i've used R a lot, but i don't know what "panel data" is sorry. maybe take a look at some of the jamovi tutorials on the community page, and see if that clears things up.

https://www.jamovi.org/community.html

alternatively, someone else might be familiar with what you mean, and might chime in.

cheers

jonathon
Jorge
Posts: 22
Joined: Thu Apr 01, 2021 8:44 pm

Re: Cross section,Time Series and Panel data (balance&unbala

Post by Jorge »

In econometric analysis, it is usual to distinguish between cross sectional data and panel data. While in a cross section we have one observation for each individual, in panel data we have many observations for each individual (one observation for each time period).

Although I am not an expert, as far as i know, jamovi don't work with panel data (at least for now).

Cheers,
Jorge
User avatar
jonathon
Posts: 2613
Joined: Fri Jan 27, 2017 10:04 am

Re: Cross section,Time Series and Panel data (balance&unbala

Post by jonathon »

oh yup. this sounds like what we call "long format" and "wide format" ...

the analyses in jamovi assume that each row is independent of every other row ... however, there are some analyses in gamlj where multiple rows are grouped for a single 'case' (long format).

cheers

jonathon
kalemguy
Posts: 3
Joined: Fri Oct 07, 2022 2:43 am

Re: Cross section,Time Series and Panel data (balance&unbala

Post by kalemguy »

Thank you for your explanations @Jorge and @Jonathon

Yes, I think it is "long format". Long format datasets stack the observations of each variable from all groups, across at all time periods into one column (https://www.aptech.com/blog/introductio ... anel-data/)

So this type of data can ONLY be analysed using GAMLj module?
However, is there any means in Jamovi able to justify N and T for this "long format (Panel)" dataset?

Thank You
Post Reply