Four-level CCM, variance components

Welcome to the forum for runmlwin users. Feel free to post your question about runmlwin here. The Centre for Multilevel Modelling take no responsibility for the accuracy of these posts, we are unable to monitor them closely. Do go ahead and post your question and thank you in advance if you find the time to post any answers!

Go to runmlwin: Running MLwiN from within Stata >> http://www.bristol.ac.uk/cmm/software/runmlwin/
Post Reply
dand1994
Posts: 2
Joined: Tue Oct 29, 2019 4:43 pm

Four-level CCM, variance components

Post by dand1994 » Wed Oct 30, 2019 6:35 pm

Hello.

I am attempting to put together a four-level CCM using the runmlwin command. As shown below, the "firm" at level 3 is nested within home countries at level 4, while the host country and industry are cross-nested.

Outcome
  • LEVEL 1: Year
  • LEVEL 2: Unit
  • LEVEL 3: Host country, Industry, Firm
  • LEVEL 4: Home country
It would look something like this:
  • Level 1: ROAyshipm = β0shipm + eyshipm
  • Level 2: β0shipm = β00hipm + r0shipm
  • Level 3: β00hipm = β000m + b00h + c00i + d00p
  • Level 4: β000m = 𝛾000 + e000m
Any help with putting this into Stata?

ChrisCharlton
Posts: 1132
Joined: Mon Oct 19, 2009 10:34 am

Re: Four-level CCM, variance components

Post by ChrisCharlton » Mon Nov 11, 2019 3:26 pm

You can find examples that replicate the MLwiN manuals using runmlwin at https://www.bristol.ac.uk/cmm/software/ ... /examples/. I would suggest looking at the example for Chapter 15 of the MCMC guide, as this covers setting up cross-classified models.

dand1994
Posts: 2
Joined: Tue Oct 29, 2019 4:43 pm

Re: Four-level CCM, variance components

Post by dand1994 » Wed Dec 04, 2019 10:00 pm

ChrisCharlton wrote:
Mon Nov 11, 2019 3:26 pm
You can find examples that replicate the MLwiN manuals using runmlwin at https://www.bristol.ac.uk/cmm/software/ ... /examples/. I would suggest looking at the example for Chapter 15 of the MCMC guide, as this covers setting up cross-classified models.
Unfortunately, however, the examples provided are not directly relatable due to the complexity of my model. Could you perhaps shed a bit more light on running the model (e.g., runmlwin command)? I am attaching a picture showing the data structure.

I have used: runmlwin s_roa cons, level6(home : cons) level5(firm: cons) level4(industry: cons) level3(host: cons) level2(unit: cons) level1(year: cons) mcmc(cc) initsprevious nopause, although I do not believe it to be forrect

Image

GeorgeLeckie
Site Admin
Posts: 428
Joined: Fri Apr 01, 2011 2:14 pm

Re: Four-level CCM, variance components

Post by GeorgeLeckie » Fri Dec 06, 2019 3:59 pm

Dear dand1994,

"I have used: runmlwin s_roa cons, level6(home : cons) level5(firm: cons) level4(industry: cons) level3(host: cons) level2(unit: cons) level1(year: cons) mcmc(cc) initsprevious nopause, although I do not believe it to be forrect"

What you have written is correct for a cross-classified model

Once you specify mcmc(cc), the ordering of the levels is irrelevant.

I hope that helps

George

Post Reply