mean
The average, together with how precise that average is.
What we have
- the mean with its standard error and confidence interval
over()and weights
Coming later
- a survey design — weights alone give the right mean and the wrong standard error; use svy: mean
How you write it
mean varlist [if] [weight] [, over(var)]
Square brackets mean optional. A vertical bar means choose one.
What it does
summarize gives you the average. mean gives you the average plus its standard error and a 95% confidence interval — the range the true average plausibly lies in.
over(sector) reports it separately for each group. Weights are respected.
Watch out. This treats your rows as a simple random sample. If the data came from a survey with a design — stratified, clustered, weighted — use svy: mean instead, or the interval will be too narrow.
Try it
Every example below was run after sysuse vira.sample02, clear.
. mean mpce
Mean estimation
Number of obs = 6,000
Design df = 5,999
| Mean Std. err. [95% conf. interval]
-----------------+--------------------------------------------------------
mpce | 3718.605 19.16605 3681.033 3756.178
The same thing in pandas and in R
pandas
# mean of ['mpce'] with linearised standard errors
R
# no R form for `mean` yet -- see docs/r-equivalents.md