How To Use Weights Aggregate Census Data Stata

Observe: On April 23, 2014, Statalist moved from an email listing to a forum, based at statalist.org.

[Engagement Prev][Engagement Next][Thread Prev][Thread Side by side][Engagement Index][Thread Index]

st: Re: Weighting by State Population: Apply iweight?

Utilise aweights - i.due east. [aw=state_pop]. If yous were to use iweights, the implied sample size and the standard errors would depend upon the arbitrary scaling of state_pop.

In this context aweights are unlike from the weights used by the BLS, etc to construct state-level statistics. What aweights do is to requite a greater weight to rates (crime, unemployment, etc) for states with big populations than to those with small populations.

Notation one potential trap, which concerns the footing for normalising the data. For some purposes you can treat each state & twelvemonth combination every bit a separate observation, and then that AZ may have a larger weight in 2005 than in 1980. This is ok if you utilize OLS and then calculate standard errors adapted for clustering by state. On the other manus, if y'all wish to use panel data procedures, these will normally insist that the weight is the aforementioned for all years for a item state (panel unit). In that case, you have to use (for example) population in 2000 as the weight variable or the boilerplate state population over the information period, which can be calculated easily using -egen-.

            Gordon Hughes g.a.hughes@ed.ac.u.k.  =====================================  I am working on running a panel data regression that compares various law-breaking rates (dependent) to unemployment and income inequality (contained). The panel consists of fifty states from 1976 to 2009. Nearly of the previous literature when providing summary statistics and OLS regression results simply state that the statistics and regressions are "weighted by country population". I am very confused on how to weight by country population. I do not recall I need to use pweight or aweight equally the data is already aggregated by the US Census and Bureau of Labor Statistics.Would I, therefore, use  the iweight command? Could create a new variable that is the per centum of the total US population that a state posses and use this as the weight?  * *   For searches and aid endeavour: *   http://www.stata.com/aid.cgi?search *   http://www.stata.com/support/statalist/faq *   http://world wide web.ats.ucla.edu/stat/stata/

Prev past Date: st: two-stage estimation of probit with kickoff-stage oprobit
Next by Date: RE: st: How to generate dummy for the following years
Previous by thread: st: 2-stage estimation of probit with outset-stage oprobit
Next past thread: st: count instances beyond a group of variables
Index(es):
- Date
- Thread