I have the following situation:
The variables I have selected for a regression contain values for the years 1999-2011, except for the binary variable "poolacc"(= does the property has a swimming pool?) for which I have only values for the year 2011. I believe that I have to copy the same values of 2011 to the other years before (1999-2010).
How can I fill in the dots for years 1999-2010 of this variable 'poolacc' with the values of this particular year 2011 in an efficient way?
Here is some data with 3 variables; the id (I generated myself) ; hhmove=year moved into house; poolacc=presence of a pool (1= Yes, 2=No).
Code:
* Example generated by -dataex-. To install: ssc install dataex clear input float id int hhmove byte poolacc 1 2001 . 2 1999 . 3 1999 . 4 1999 . 5 1999 . 6 2000 . 7 1999 . 8 2000 . 9 2000 . 10 1999 . 11 2000 . 12 2001 . 13 2000 . 14 1999 . 15 2001 . 16 2001 . 17 1999 . 18 1999 . 19 1999 . 20 1999 . 21 1999 . 22 2000 . 23 2001 . 24 2000 . 25 2000 . 26 1999 . 27 2000 . 28 1999 . 29 1999 . 30 1999 . 31 1999 . 32 2000 . 33 2001 . 34 1999 . 35 2000 . 36 2000 . 37 2000 . 38 2001 . 39 2000 . 40 1999 . 41 2001 . 42 1999 . 43 2001 . 44 2001 . 45 2000 . 46 2001 . 47 2001 . 48 2001 . 49 2000 . 50 1999 . 51 1999 . 52 1999 . 53 2000 . 54 1999 . 55 1999 . 56 2001 . 57 1999 . 58 2001 . 59 1999 . 60 1999 . 61 2000 . 62 1999 . 63 1999 . 64 2001 . 65 1999 . 66 2000 . 67 1999 . 68 2000 . 69 1999 . 70 1999 . 71 2001 . 72 2000 . 73 2000 . 74 2000 . 75 2001 . 76 1999 . 77 2001 . 78 2000 . 79 1999 . 80 1999 . 81 2000 . 82 1999 . 83 2001 . 84 1999 . 85 2001 . 86 2001 . 87 2000 . 88 1999 . 89 2000 . 90 2000 . 91 2000 . 92 2000 . 93 2000 . 94 2000 . 95 2001 . 96 2000 . 97 2001 . 98 2000 . 99 1999 . 100 2000 . end
Code:
bysort id (hhmove): replace poolacc = poolacc[_n -1] if poolacc ==. & hhmove < 2011
Code:
by id (hhmove), sort: replace poolacc= poolacc[_n-1] if poolacc>= . & hhmove<2011
Does anyone know which code I need to put in to solve this issue?
Thanks in advance!
0 Response to Replace missing values by (copying) non-missing values
Post a Comment