In this code, I want to create and store an interpolation function for
each date. It seems that data.table saves the same interpolation
function (for the last date) for all three dates. How can this problem
be fixed?
Thanks,
Naresh
mydt <- list(data.table(date = as.IDate("2025-01-01"), x = c(1, 2, 3), y = c(3,
4, 5)), data.table(date = as.IDate("2025-01-15"), x = c(1, 2, 3), y = c(9, 8,
7)), data.table(date = as.IDate("2025-01-31"), x = c(1, 2, 3), y = c(10, 12,
9)))
mydt <- rbindlist(mydt)
myapprox <- mydt[, list(yfunc = list(approxfun(x, y, rule = 2))), by =
list(date)]
testdt <- data.table(date = as.IDate(c("2025-01-01", "2025-01-15",
"2025-01-31")), x = 2)
testdt <- merge(testdt, myapprox, by = "date")
testdt[, let(yapprox = yfunc[[1]](x)), by = .I]
testdt <- merge(testdt, mydt, by = c("date", "x"))
testdt
Key: <date, x>
date x yfunc yapprox y
<IDat> <num> <list> <num> <num>
1: 2025-01-01 2 <function[1]> 12 4
2: 2025-01-15 2 <function[1]> 12 8
3: 2025-01-31 2 <function[1]> 12 12
Using split function, it is possible to get the desired result.
mydtsplit <- split(mydt, f = mydt$date)
splitapprox <- lapply(mydtsplit, function(dt) dt[, approxfun(x, y,
rule = 2)])
newdt <- data.table(date = as.IDate(c("2025-01-01", "2025-01-15",
"2025-01-31")), x = 2)
newdt[, let(yfunc = splitapprox[as.character(date)]), by = .I]
newdt[, let(yapprox = yfunc[[1]](x)), by = .I]
newdt <- merge(newdt, mydt, by = c("date", "x"))
newdt
Key: <date, x>
date x yfunc yapprox y
<IDat> <num> <list> <num> <num>
1: 2025-01-01 2 <function[1]> 4 4
2: 2025-01-15 2 <function[1]> 8 8
3: 2025-01-31 2 <function[1]> 12 12
[[alternative HTML version deleted]]
______________________________________________
[email protected] mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide https://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
