The Transition Probability Matrix: Rows that Sum to 1, Absorbing States and Two-Step Paths
This lesson is part of the course
In this lesson
You divide the transition counts by their row totals to get the probability matrix P, add the absorbing Conversion and Null rows, and use MMULT to get two- and three-step probabilities.
Buy the course once for $9.99 to unlock every lesson and lab, the final exam and the certificate – lifetime access, no subscription.
Lessons in this course
- From Customer Journeys to Transition Counts: Start, Channel Steps, Conversion and Null Locked
- The Transition Probability Matrix: Rows that Sum to 1, Absorbing States and Two-Step Paths Locked
- Conversion Probability from Start: Step-by-Step Convergence and the Exact (I − Q)⁻¹ Solution Locked
- Removal Effects: What Happens to Conversions when a Channel Disappears Locked
- Attribution Shares: Normalising Removal Effects into Credited Conversions and Revenue Locked
- Markov vs Last Click vs Linear: Who Gains, Who Loses (with a Chart) Locked
- Budget Implications: Cost per Attributed Conversion, ROAS and a Naive Reallocation Locked
- Sensitivity & Limitations: How Stable Is the Credit, and What Markov Attribution Cannot Tell You Locked