Create a time series dummy variable.

`time.dummy <- function(x, dates1, dates2=NULL)`

`x`

: The time series for which you want to create a matching dummy variable.`dates1`

and`dates2`

: See below.

If `dates1`

is a single date and `dates2`

is `NULL`

, this function creates a dummy variable equal to `TRUE`

for that single date.

If `dates1`

is a list of dates and `dates2`

is `NULL`

, creates a dummy variable equal to `TRUE`

on each date in `dates1`

.

If `dates2`

is a single date and `dates2`

is a single date, creates a dummy variable equal to `TRUE`

for all dates in that range.

If `dates1`

is “start”, sets `dates1`

to the first observation.

If `dates2`

is “end”, sets `dates2`

to the last observation.

```
time.dummy(y, c(2004,7)) # One observation is TRUE
time.dummy(y, c(2004,7), c(2005,8)) # 14 observations are TRUE
time.dummy(y, "start", c(2005,8))
time.dummy(y, c(2004,7), "end")
time.dummy(y, "start", "end") # Legal, but doesn't make any sense
time.dummy(y, list(c(2003,8), c(2004,9), c(2009, 12))) # Three observations are TRUE
```