{% include "header.html" %}

<div class="controlpanel text-center">
  <h2>Minimum Viable Model Estimator</h2>
</div>

<div class="input-group" style="margin: 0 auto; width: 680px;">
<p>
This application allows you to estimate the minimum required performance characteristics
of a binary classification model. You need to specify the expected costs and benefits of
making decisions on the basis of the model. Including the number of cases uyou expect to
apply it to in a given period of time, and the base rate of the event you are attemting
to predict.
</p>
<p>
Please note that costs should be represented as negative numbers in the cost/benefit matrix.
</p>
<p>
If your problem requires an intervention with stochastic outcomes we request that you supply
the following alternative values for estimation:

<ul>
<li>Intervention Cost: The dollar cost of making the intervention (non-negative number)
</li>
<li>Success Probability: The probability that your intervention will succeed.</li>
<li>Backfire Probability: The probability that your intervention will have the opposite effect (i.e. Cause a customer to churn who otherwise would not have.)</li>
<li>Success Impact:  The dollar impact of a successful intervention.</li>
<li>Backfire Impact: The dollar impact of a backfire event (please use a negative number for negative impact).</li>

</ul>
</p>

</div>

{% include "footer.html" %}