我的原始数据是一个公司代码(stkcd)对应很多个年份(2008-2018),每个年份下面又对应很多个具体数据(personid),对于缺失值的处理我有两个想法,一是只要含有缺失值,则将整个公司各年的数据全部删除,二是只要含有缺失值,则将缺失值对应的年份下的所有数据都删除,比如如下数据中倒数第二行有一个缺失值,针对这个缺失值我想方法一是可以将stkcd为000002的数据全部删除,方法二是仅将stkcd为000002,year为2012的数据删除,求教大神用stata如何分别实现这两个目的,十分感谢!
- * Example generated by -dataex-. To install: ssc install dataex
- clear
- input str6 stkcd int year long personid str3 gender byte(age degree) int tenure
- "000002" 2008 309304 "男" 38 4 94
- "000002" 2008 3038744 "男" 37 3 41
- "000002" 2008 3043528 "男" 43 4 57
- "000002" 2008 3053716 "男" 43 4 94
- "000002" 2008 3054222 "女" 44 4 14
- "000002" 2008 30142574 "男" 41 4 52
- "000002" 2008 30156222 "男" 42 4 14
- "000002" 2008 30156240 "男" 39 4 72
- "000002" 2008 30156346 "男" 41 4 50
- "000002" 2009 309304 "男" 39 4 106
- "000002" 2009 3038744 "男" 38 3 53
- "000002" 2009 3043528 "男" 44 4 69
- "000002" 2009 3053716 "男" 44 4 20
- "000002" 2009 3054222 "女" 45 4 26
- "000002" 2009 30130814 "男" 37 3 9
- "000002" 2009 30142574 "男" 42 4 64
- "000002" 2009 30156222 "男" 43 4 26
- "000002" 2009 30156240 "男" 40 4 84
- "000002" 2009 30156346 "男" 42 4 62
- "000002" 2010 309304 "男" 40 4 118
- "000002" 2010 3038744 "男" 39 3 65
- "000002" 2010 3043528 "男" 45 4 81
- "000002" 2010 3053716 "男" 45 4 32
- "000002" 2010 3054222 "女" 46 4 38
- "000002" 2010 30130814 "男" 38 3 21
- "000002" 2010 30142574 "男" 43 4 76
- "000002" 2010 30156222 "男" 44 4 38
- "000002" 2010 30156240 "男" 41 4 96
- "000002" 2010 30156346 "男" 43 4 74
- "000002" 2011 309304 "男" 41 4 130
- "000002" 2011 3043528 "男" 46 4 93
- "000002" 2011 3044616 "男" 56 4 9
- "000002" 2011 3052888 "男" 43 5 9
- "000002" 2011 3053716 "男" 46 4 9
- "000002" 2011 3054222 "女" 47 4 50
- "000002" 2011 30130814 "男" 39 3 33
- "000002" 2011 30142574 "男" 44 4 88
- "000002" 2011 30151046 "男" 41 4 11
- "000002" 2011 30156222 "男" 45 4 50
- "000002" 2011 30156346 "男" 44 4 86
- "000002" 2012 309304 "男" 42 4 144
- "000002" 2012 3043528 "男" 47 4 105
- "000002" 2012 3044616 "男" 57 4 21
- "000002" 2012 3052888 "男" 44 5 21
- "000002" 2012 3053716 "男" 48 4 142
- "000002" 2012 3054222 "女" 49 4 62
- "000002" 2012 30130814 "男" 40 3 45
- "000002" 2012 30142574 "男" 46 4 100
- "000002" 2012 30151046 "男" 42 4 .
- "000002" 2012 30156222 "男" 47 4 62
- end
复制代码