GARCH commands for panel data?

[Marta ArespaC]

Dear all,

I work with an unbalanced panel data of 100 individuals (CPI-U subcategories) and monthly data from 1950 to 2020. I want to study inflation volatility using GARCH models.
This has been done in papers like this and this (see citations below), back in 2010 and 2016, but authors were using RATS, that I do not have. They indicated me the routine for RATS was available here.
I imagine Stata has a way to do it, but I did not find it in the panel data manual, help arch or forums...

How could I use Stata to regress an

arch lninf_a, ar(2) ma(0) arch(1) garch(1)

where lnif_a is

ln(inflation)-ln(inflation_[n-12])

if lninf_a is available for the 100 individuals in the panel? I do not want a regression for each panel, but one for all.

Thank you in advance for your help.

1. Lee, Jim, 2010. "The link between output growth and volatility: Evidence from a GARCH model with panel data," Economics Letters, Elsevier, vol. 106(2), pages 143-145, February.
2. Jim Lee & Harold Glenn A. Valera, 2016. "Price transmission and volatility spillovers in Asian rice markets: Evidence from MGARCH and panel GARCH models," The International Trade Journal, Taylor & Francis Journals, vol. 30(1), pages 14-32, January.

[Marta ArespaC]

I understand that a panel GARCH is like a DVECH with common dynamics in variance and covariance processes. Just in case this helps to get an answer...

[George Ford]

mgarch?

https://www.youtube.com/watch?v=NrodEW1iAHc

[Marta ArespaC]

Dear George Ford,

Thank you so much for giving an answer... I watched the video you linked.

However, I am not convinced mgarch estimates what I would like to... For instance, I would like to have only one coefficient for lagged-dependent (lag-y) variable, instead of one for each of the 100 series I work with...
I would like to replicate/estimate (for my dataset) model A in Cermeño and Grier (2003), with (constant) fixed effects. See equations 1 to 4 in pages 4 and 5 of the cited paper.

Any help is so welcome.

[George Ford]

I don't think Stata can do it with a canned program. R and Eviews can, I think.

[Edson Vengesai]

@ Marta, did you get help on the Panel GARCH? Kindly share if you found a way out.