mrwawa
01-19-2011, 01:45 PM
Hi all,
I am working with a demographic data set that consists of three separate transition matrices. I have simulated growth from an initial vector N0 through 50 time steps with 5000 replication under three different probability scenarios for selecting the matrix at t+1. As an example, one scenario might have equal probabilities, the next (.5,.25,.25), and the final scenario (.25,.5.,25). My response is a set of 5000 values for each of the three scenarios. The responses represent the stochastic population growth rate under the three separate scenarios. Under the three scenarios, the means of the 5000 values are different, but I am wondering how I can actually test for differences. Based on theory and the results, the means are log-normally distributed, but not sure what the degrees of freedom would be.
Does anyone know how to do this?
Thanks for any information,
Wade
I am working with a demographic data set that consists of three separate transition matrices. I have simulated growth from an initial vector N0 through 50 time steps with 5000 replication under three different probability scenarios for selecting the matrix at t+1. As an example, one scenario might have equal probabilities, the next (.5,.25,.25), and the final scenario (.25,.5.,25). My response is a set of 5000 values for each of the three scenarios. The responses represent the stochastic population growth rate under the three separate scenarios. Under the three scenarios, the means of the 5000 values are different, but I am wondering how I can actually test for differences. Based on theory and the results, the means are log-normally distributed, but not sure what the degrees of freedom would be.
Does anyone know how to do this?
Thanks for any information,
Wade