I had some excel spreadsheets, containing A large amount of data, A column represents the update time (from 2013-1-18 to 2013-1-18), B and C column represents provinces and cities and locations, D column represents data column (format below)

1. When the data loss & lt; H=5 (this is less than five data loss, including five), then use linear interpolation (such as TREND function) to supplement data,

2. When the data loss & gt; 5 h (that is, the loss of more than five data), then on the basis of data loss in the day, should use a day after the day before and the relative time point data of calculating mean,

Teacher, please help solve, thank you