Real-time event-clock forecast

The real-time benchmark used in the accompanying working paper: estimate the current information intensity from a trailing window and scale it to the forecast horizon, $$\widehat A^{rt}(h) = \frac{RV(\text{trailing } k \text{ obs.})} {\text{calendar span in days}} \times h.$$

Usage

event_clock_forecast(
  x,
  at = NULL,
  horizon,
  trailing = ec_default_params()$trailing,
  sample_every = ec_default_params()$sample_every,
  clip = NULL
)

Arguments

x

An event_prices object (see as_event_prices()), or a data.frame coercible to one.

at

Valuation date/time at which the forecast is made (default: last observation).

horizon

Forecast horizon(s): either numeric (days) or Date/POSIXct horizon dates (converted to days from at). Names are used as labels.

trailing

Integer, number of trailing observations in the estimation window (default 40, the paper's headline choice; 20 and 60 are common robustness settings).

sample_every

Integer, use every k-th observation (sparse-sampling robustness; default 1).

clip

Numeric length-2 clipping bounds for q before the log-odds transform; defaults to the bounds stored in x.

Value

A tibble with columns market_id, at, horizon, horizon_days, trailing, n_incr, n_gaps, max_gap_days, and A_forecast. See event_clock() for the gap diagnostics.

Examples

data(us2016)
ep <- as_event_prices(us2016, time = "date", price = "trump")
event_clock_forecast(ep,
  at = as.Date("2016-10-10"),
  horizon = c(`1W` = 7, `2W` = 14)
)
#> # A tibble: 2 × 9
#>   market_id at         horizon horizon_days trailing n_incr n_gaps max_gap_days
#>   <chr>     <date>     <chr>          <dbl>    <int>  <int>  <int>        <dbl>
#> 1 NA        2016-10-10 1W                 7       40     39      0            1
#> 2 NA        2016-10-10 2W                14       40     39      0            1
#> # ℹ 1 more variable: A_forecast <dbl>