* 为每行创建一个新变量来储存非重复字符串的数量
gen unique_count = 0
foreach var of varlist n1 n2 n3 n4 {
* 这里假设你的数据中的字符串变量是n1, n2, n3, 和 n4
quietly: by year id (n`var'), sort: replace unique_count = unique_count + cond(n`var' != n1 & n`var' != n2 & n`var' != n3 & n`var' != n4, 1, 0)
}
* 为每个year和id组合加总unique_count
egen total_unique_count = total(unique_count), by(year id)