Calculate variable time-step net transition rates

Here we provide an example of a model that can be used to run multiple time step simulations when the time-period matrix cannot be decomposed to a multiple-step transition matrix because it is not an ergodic matrix, i.e. Markov chain matrix.

For matrices that cannot be decomposed into multiple-step transition matrices, this models calculates first gross transition rates for a time-period and a specified time-step and then derives variable net transition rates, producing, as a result, a unique transition matrix for each model iteration. Output transition matrices are saved into files that can be loaded in a simulation with the same number of time-steps.

This model, calculate time-step transition rates.egoml, and the model that runs the simulation with the resulting matrices, savassi1996 2003 in annual steps.egoml, are in guidebook dinamica\models\variable time step net transition rates\simulate 2003 in annual time steps, part of guidebook_dinamica.7z. Their input maps, from the urban dynamics example of the Savassi neighborhood in Belo Horizonte, are in the originals folder next to it.

— Britaldo Soares Filho 2011/03/11 13:16

See also: Varying Parameters in a Simulation shows how to use a different transition matrix at each step of a simulation.

☞Next: Receiver Operating Characteristic (ROC) suite


This documentation was generated with the help of large language models (LLMs) and may contain errors. Please verify critical information.