Skip to main content
Version: 3.13

compare_models

Compare the performance of first touch, last touch, linear touch, Markov model, Shapley value and logistic regression in a simulated traffic allocation problem based on real data.

Parameters

PARAMETERTYPEDEFAULTDESCRIPTION
df_pathsdata.framedata.frame including customer journeys
var_pathstrNonename of the column containing paths
var_convstrNonename of the column containing total conversions
var_nullstrNonename of the column containing total paths that do not lead to conversion
cha_sepstr">"separator between channels
channel_conv_namestrNonehow conversion state is identified in df_paths
df_ctrdata.frameNonedata.frame with click-through rates for channels expressed in number of clicks
perc_realldouble0.10Percentage of traffic to reallocate
perc_sampledouble0.10Percentage of sampled customer journeys
max_nsimint10000Maximum number of sampled customer journeys
min_perc_trafficdouble0.005Minimum percentage of traffic equally split among each channel
niterint10Number of iterations
flg_extra_pathbool1If = 1, extra paths will be added to Data
max_markov_orderint3all the Markov models of order <= max_markov_order will be computed
seedint1234567random seed
verboseint1If = 1, some information is printed during the execution

Output

OUTPUTTYPEDESCRIPTION
performancedata.frameperformance in term of conversion rate reached by each attribution model
allocationdata.frametraffic allocation suggested by each attribution model