Geometric adstock weights

The finite geometric kernel \(w_i = \theta^i\) for lags \(i = 0, 1, \ldots, L-1\). adstock_geometric() calls this internally when max_lag is finite; it is exported because the weights themselves are often what you want to plot or report.

Usage

adstock_weights(max_lag, decay, normalise = TRUE)

Arguments

max_lag

Number of periods the kernel spans, including the current period. max_lag = 1 means no carryover.

decay

Geometric decay coefficient in [0, 1]. See decay_from_half_life().

normalise

Should the weights sum to 1? See the Normalisation section of adstock_geometric() for what this does to the interpretation of a downstream coefficient.

Value

A numeric vector of length max_lag, ordered from the current period to the most distant lag.

See also

adstock_geometric(), adstock_weights_weibull()

Examples

adstock_weights(max_lag = 5, decay = 0.5)
#> [1] 0.51612903 0.25806452 0.12903226 0.06451613 0.03225806

# Unnormalised: the current period always carries weight 1
adstock_weights(max_lag = 5, decay = 0.5, normalise = FALSE)
#> [1] 1.0000 0.5000 0.2500 0.1250 0.0625

# Normalised weights sum to 1, so the series level is preserved
sum(adstock_weights(max_lag = 10, decay = 0.7))
#> [1] 1