Version: 3.20.1

mmm_attribution

R

D_variables must include prior_weight. The value is applied to each raw input signal before saturation and scoring.

prior_weight: business weight assigned to each raw signal. A common way to estimate it is conversions / touchpoints, but it can also be set from domain knowledge or historical benchmarks.

library(ChannelAttributionPro)

token <- "yourtoken"

Data <- read.csv("https://app.channelattribution.io/data/data_mmm_3.csv")

D_variables <- data.frame(
  variable = c(
    "direct_searches",
    "facebook_impressions",
    "facebook_clicks",
    "google_impressions",
    "google_clicks",
    "tv"
  ),
  channel = c(
    "direct",
    "facebook_ads",
    "facebook_ads",
    "google_ads",
    "google_ads",
    "tv"
  ),
  measure = c(
    "direct_searches",
    "impressions",
    "clicks",
    "impressions",
    "clicks",
    "views"
  ),
  prior_weight = c(
    0.50,   # direct search
    0.015,  # Facebook impressions
    0.21,   # Facebook clicks
    0.015,  # Google impressions
    0.25,   # Google clicks
    0.02    # TV views / GRP-like signal
  ),
  stringsAsFactors = FALSE
)

target <- "conversions"

res <- mmm_attribution(
  Data = Data,
  D_variables = D_variables,
  target = target,
  model = "linear", # "linear", "reward", or "copula"
  max_p = 12,
  seed = 1234567,
  verbose = 1,
  password = token
)

print(head(res))

Using a different internal model:

res_reward <- mmm_attribution(
  Data = Data,
  D_variables = D_variables,
  target = target,
  model = "reward",
  max_p = 12,
  seed = 1234567,
  password = token
)

res_copula <- mmm_attribution(
  Data = Data,
  D_variables = D_variables,
  target = target,
  model = "copula",
  max_p = 12,
  nsim = 1000,
  seed = 1234567,
  password = token
)

R​

R