r - How to get values with same time with previous days same value -

how replace nas in var_1 same previous days value

eg.row 4 has missing value - replace value/na same time in previous day(row 1).

similarly row 7 row 3, row 10 row 9 , row13 row 11. can replace na previous value zoo-na.lacf

id  date                var var_1 1   21-01-2014 00:15    22  22 2   21-01-2014 00:30    33  33 3   21-01-2014 00:45    13  13 4   22-01-2014 00:15    na  22 5   22-01-2014 00:30    22  22 6   22-01-2014 00:35    54  54 7   22-01-2014 00:45    na  13 8   23-01-2014 00:15    23  23 9   25-01-2104 01:00    34  34 10  26-01-2104 01:00    na  34 11  27-01-2104 02:00    2   2 12  27-01-2104 03:00    21  21 13  28-01-2104 02:00    na  2 

one way use format time, , split on that. use na.locf fill nas , rbind together, i.e.

library(zoo)  do.call(rbind, c(lapply(split(df, format(df$date, format = '%h:%m')), function(i) {                                      i$var <- na.locf(i$var);                                   }), make.row.names = false)) 

which gives,

 id                date var var_1 1   1 2014-01-21 00:15:00  22    22 2   4 2014-01-22 00:15:00  22    22 3   8 2014-01-23 00:15:00  23    23 4   2 2014-01-21 00:30:00  33    33 5   5 2014-01-22 00:30:00  22    22 6   6 2014-01-22 00:35:00  54    54 7   3 2014-01-21 00:45:00  13    13 8   7 2014-01-22 00:45:00  13    13 9   9 2104-01-25 01:00:00  34    34 10 10 2104-01-26 01:00:00  34    34 11 11 2104-01-27 02:00:00   2     2 12 13 2104-01-28 02:00:00   2     2 13 12 2104-01-27 03:00:00  21    21 

you can order on id if per usualdf[with(df, order(id)),]

wiki