Markov Transition Matrix

A Markov transition matrix is a square matrix specifying the frequency with which the values in a time series transition from one state to another when moving from one time step to the next.

To construct a Markov transition matrix, Windographer divides the range of values exhibited by a time series into N bins that serve as possible 'states' for that variable. Then it steps through the time series counting the number of times that the value transistions from state i in one time step to state j in the next time step.

Take for example the time series that appears below. It has a time step of 10 minutes, varies in value between zero and 100%, and covers an 18-month period, part of which appears in this graph:

By dividing the range [0%, 100%] into 25 bins representing 25 states, Windographer calculated the following Markov transition matrix for this sample time series:

Initial State Final State
0%-4% 4%-8% 8%-12% 12%-16% 16%-20% 20%-24% 24%-28% 28%-32% 32%-36% 36%-40% 40%-44% 44%-48% 48%-52% 52%-56% 56%-60% 60%-64% 64%-68% 68%-72% 72%-76% 76%-80% 80%-84% 84%-88% 88%-92% 92%-96% 96%-100%
0%-4% 0.861 0.125 0.010 0.002 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
4%-8% 0.126 0.702 0.147 0.019 0.005 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
8%-12% 0.005 0.130 0.694 0.146 0.017 0.004 0.002 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
12%-16% 0.001 0.014 0.152 0.621 0.168 0.028 0.009 0.003 0.001 0.001 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
16%-20% 0.001 0.005 0.021 0.134 0.607 0.180 0.034 0.012 0.003 0.002 0.001 0.001 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
20%-24% 0.000 0.001 0.006 0.024 0.154 0.596 0.163 0.038 0.012 0.004 0.002 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
24%-28% 0.000 0.001 0.002 0.010 0.031 0.153 0.581 0.168 0.035 0.010 0.005 0.002 0.001 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
28%-32% 0.000 0.000 0.001 0.004 0.013 0.036 0.167 0.545 0.172 0.039 0.012 0.006 0.004 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
32%-36% 0.000 0.000 0.000 0.003 0.007 0.017 0.046 0.166 0.519 0.171 0.037 0.018 0.006 0.005 0.002 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
36%-40% 0.000 0.000 0.000 0.000 0.002 0.006 0.016 0.038 0.158 0.539 0.169 0.047 0.015 0.004 0.003 0.000 0.001 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
40%-44% 0.000 0.000 0.000 0.000 0.000 0.002 0.008 0.011 0.039 0.157 0.538 0.173 0.046 0.016 0.005 0.005 0.001 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
44%-48% 0.000 0.000 0.000 0.000 0.000 0.001 0.001 0.007 0.015 0.043 0.163 0.533 0.170 0.040 0.013 0.007 0.002 0.001 0.001 0.000 0.000 0.000 0.000 0.000 0.000
48%-52% 0.000 0.000 0.000 0.000 0.000 0.000 0.001 0.004 0.006 0.014 0.043 0.155 0.552 0.163 0.038 0.013 0.006 0.003 0.002 0.000 0.000 0.000 0.000 0.000 0.000
52%-56% 0.000 0.000 0.000 0.000 0.000 0.000 0.001 0.002 0.004 0.006 0.013 0.036 0.155 0.548 0.168 0.042 0.014 0.006 0.003 0.001 0.000 0.000 0.000 0.000 0.000
56%-60% 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.005 0.006 0.011 0.036 0.144 0.576 0.166 0.038 0.012 0.003 0.002 0.001 0.000 0.000 0.000 0.000
60%-64% 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.001 0.002 0.005 0.006 0.014 0.034 0.161 0.552 0.178 0.033 0.008 0.005 0.001 0.000 0.001 0.000 0.000
64%-68% 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.001 0.006 0.009 0.013 0.040 0.174 0.530 0.171 0.038 0.010 0.005 0.002 0.001 0.000 0.000
68%-72% 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.001 0.002 0.003 0.004 0.013 0.030 0.157 0.568 0.177 0.029 0.013 0.002 0.000 0.000 0.000
72%-76% 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.001 0.001 0.004 0.004 0.013 0.032 0.160 0.576 0.161 0.032 0.012 0.002 0.000 0.000
76%-80% 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.002 0.003 0.006 0.014 0.034 0.148 0.601 0.156 0.028 0.006 0.002 0.000
80%-84% 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.001 0.002 0.006 0.014 0.029 0.149 0.624 0.147 0.025 0.004 0.001
84%-88% 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.001 0.000 0.001 0.003 0.004 0.010 0.028 0.150 0.656 0.130 0.014 0.002
88%-92% 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.001 0.002 0.003 0.011 0.021 0.133 0.694 0.126 0.010
92%-96% 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.001 0.003 0.007 0.018 0.136 0.727 0.107
96%-100% 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.002 0.013 0.110 0.874

That transition matrix shows that if the value of the time series is in the 0-4% range in some time step, there is:

Similarly, it shows that if the value of the time series is in the 68-72% range in some time step, there is:

Windographer uses a Markov transition matrix as part of the Markov-based reconstruction mechanism.

See also

Markov-based reconstruction mechanism


Written by: Tom Lambert
Contact: windographer.support@ul.com
Last modified: January 7, 2019