fit_spend_exposure

This function is called in <code>robyn_engineering()</code>. It uses
the Michaelis-Menten function to fit the nonlinear model. Fallback
model is the simple linear model <code><a href="/link/lm()?package=Robyn&version=3.11.1" data-mini-rdoc="Robyn::lm()">lm()</a></code> in case the nonlinear
model is fitting worse. A bad fit here might result in unreasonable
model results. Two options are recommended: Either splitting the
channel into sub-channels to achieve better fit, or just use
spend as <code>paid_media_vars</code>

Semi-Automated Marketing Mix Modeling (MMM) aiming to reduce human bias by means of ridge regression and evolutionary algorithms, enables actionable decision making providing a budget allocation and diminishing returns curves and allows ground-truth calibration to account for causation.

Bernardo Lares

Robyn

Semi-Automated Marketing Mix Modeling (MMM) from Meta Marketing
Science

Gufeng Zhou

Leonel Sentana

Igor Skokan

Meta Platforms, Inc. 

fit_spend_exposure function

<dl><dt>dt_spendModInput</dt>
<dd>data.frame. Containing channel spends and
exposure data.</dd>
<dt>mediaCostFactor</dt>
<dd>Numeric vector. The ratio between raw media
exposure and spend metrics.</dd>
<dt>paid_media_var</dt>
<dd>Character. Paid media variable.</dd></dl>

Arguments

This function is called in <code>robyn_engineering()</code>. It uses
the Michaelis-Menten function to fit the nonlinear model. Fallback
model is the simple linear model <code><a href='https://rdrr.io/r/stats/lm.html'>lm()</a></code> in case the nonlinear
model is fitting worse. A bad fit here might result in unreasonable
model results. Two options are recommended: Either splitting the
channel into sub-channels to achieve better fit, or just use
spend as <code>paid_media_vars</code>

Fit a nonlinear model for media spend and exposure — fit_spend_exposure

<dl>

<dt>dt_spendModInput</dt>
<dd>data.frame. Containing channel spends and
exposure data.</dd>


<dt>mediaCostFactor</dt>
<dd>Numeric vector. The ratio between raw media
exposure and spend metrics.</dd>


<dt>paid_media_var</dt>
<dd>Character. Paid media variable.</dd>

</dl>

fit_spend_exposure: Fit a nonlinear model for media spend and exposure

Description

Usage

Value

Arguments