tk_make_future_timeseries: Make a future time series from an existing time series
In timekit: A Collection of Tools for Working with Time Series in R

Description Usage Arguments Details Value See Also Examples

Make a future time series from an existing time series

1 2	tk_make_future_timeseries(idx, n_future, inspect_weekdays = FALSE, inspect_months = FALSE, skip_values = NULL, insert_values = NULL)

`idx`	A vector of dates
`n_future`	Number of future observations
`inspect_weekdays`	Uses a logistic regression algorithm to inspect whether certain weekdays (e.g. weekends) should be excluded from the future dates. Default is `FALSE`.
`inspect_months`	Uses a logistic regression algorithm to inspect whether certain days of months (e.g. last two weeks of year or seasonal days) should be excluded from the future dates. Default is `FALSE`.
`skip_values`	A vector of same class as `idx` of timeseries values to skip.
`insert_values`	A vector of same class as `idx` of timeseries values to insert.

tk_make_future_timeseries returns a time series based on the input index frequency and attributes.

The argument n_future determines how many future index observations to compute.

The inspect_weekdays and inspect_months arguments apply to "daily" (scale = "day") data (refer to tk_get_timeseries_summary() to get the index scale). The inspect_weekdays argument is useful in determining missing days of the week that occur on a weekly frequency such as every week, every other week, and so on. It's recommended to have at least 60 days to use this option. The inspect_months argument is useful in determining missing days of the month, quarter or year; however, the algorithm can inadvertently select incorrect dates if the pattern is irratic. For example, some holidays do not occur on the same day of each month, and as a result the incorrect day may be selected in certain years. It's recommended to always review the date results to ensure the future days match the user's expectations. It's recommended to have at least two years of days to use this option.

The skip_values and insert_values arguments can be used to remove and add values into the series of future times. The values must be the same format as the idx class. The skip_values argument useful for passing holidays or special index values that should be excluded from the future time series. The insert_values argument is useful for adding values back that the algorithm may have excluded.

A vector containing future dates

tk_index(), tk_get_timeseries_summary(), tk_get_timeseries_signature()

library(tidyquant)
library(timekit)

# Basic example
idx <- c("2016-01-01 00:00:00",
         "2016-01-01 00:00:03",
         "2016-01-01 00:00:06") %>%
    ymd_hms()
# Make next three dates in series
idx %>%
    tk_make_future_timeseries(n_future = 3)


# Create index of days that FB stock will be traded in 2017 based on 2016 + holidays
FB_tbl <- FANG %>% filter(symbol == "FB")
holidays <- c("2017-01-02", "2017-01-16", "2017-02-20",
              "2017-04-14", "2017-05-29", "2017-07-04",
              "2017-09-04", "2017-11-23", "2017-12-25") %>%
    ymd()
# Remove holidays with skip_values, and remove weekends with inspect_weekdays = TRUE
FB_tbl %>%
    tk_index() %>%
    tk_make_future_timeseries(n_future         = 366,
                              inspect_weekdays = TRUE,
                              skip_values      = holidays)

# Works with regularized indexes as well
c(2016.00, 2016.25, 2016.50, 2016.75) %>%
    tk_make_future_timeseries(n_future = 4)

# Works with zoo yearmon and yearqtr too
c("2016 Q1", "2016 Q2", "2016 Q3", "2016 Q4") %>%
    as.yearqtr() %>%
    tk_make_future_timeseries(n_future = 4)