I'm running into trouble here with something that should be easy, and I'm not sure why it isn't working. I have a dataframe modis.raw2, which has one column, day with all of the dates in the timeseries I am looking at. The other column, sr.og has values, some of which are NA.
I am trying to use lowess to smooth my data and fill in the NA values, but I keep getting an error:
Error in `$<-.data.frame`(`*tmp*`, sr.smooth, value = c(0.631508837997043, :
replacement has 50 rows, data has 128
I'm really not sure what is going on. Any ideas? Here is a sample dataset and my code.
modis.raw2 = structure(list(day = c(86, 87, 88, 89, 90, 91, 92, 93, 94, 95,
96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108,
109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121,
122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134,
135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147,
148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160,
161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173,
174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186,
187, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199,
200, 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, 212,
213), sr.og = c(0.636095833333333, NA, NA, NA, NA, 0.665508333333333,
NA, 0.67995, NA, 0.616720833333333, 0.695445833333333, 0.70355,
0.687008333333333, NA, NA, 0.6801125, 0.662941666666667, 0.688133333333333,
NA, NA, NA, NA, 0.441654166666667, 0.708595833333333, 0.698195833333333,
0.696716666666667, 0.668525, 0.622370833333333, 0.632875, 0.617458333333333,
NA, NA, NA, NA, NA, 0.64325, NA, 0.5615375, 0.547720833333333,
NA, 0.551554166666667, 0.544858333333333, 0.539941666666667,
NA, NA, 0.469941666666667, NA, 0.478479166666667, NA, 0.336470833333333,
0.3274375, 0.350295833333333, 0.288670833333333, 0.131925, 0.2291625,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, 0.0913333333333333,
NA, 0.1071125, 0.1076, 0.100704166666667, 0.1160625, NA, 0.101416666666667,
0.152929166666667, 0.1267625, NA, NA, NA, NA, NA, NA, NA, 0.1174125,
0.217633333333333, NA, NA, NA, NA, 0.225133333333333, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, 0.149670833333333, 0.154854166666667,
0.239479166666667, NA, NA, NA, NA, NA, NA, 0.153670833333333,
NA, 0.079125, NA, NA, NA, 0.208341666666667, NA, 0.102675, NA,
NA, NA, NA, NA, NA, NA)), .Names = c("day", "sr.og"), row.names = c(NA,
-128L), class = c("tbl_df", "tbl", "data.frame"))
sr.loess = loess(sr.og ~ day, data=modis.raw2, span=0.3)
complete <- tibble(day=modis.raw2$day)
complete$sr.smooth = stats::predict(sr.loess, data=complete)
The predict() function uses a parameter named newdata=, not data=. You were just extracting the fitted values here. Try
complete$sr.smooth = stats::predict(sr.loess, newdata=complete)