-
Notifications
You must be signed in to change notification settings - Fork 0
/
readme.txt
15 lines (9 loc) · 836 Bytes
/
readme.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
This code provide the mcmc samples for all movement diffusion parameters and switching parameters.
User can make inference based on the real dataset by uncomment "source('Data_real.r')" in the Setup.r file. Or make inference for the simulation data by uncomment "source('Data_sim.r')" in the Setup.r file.
In order to run the code, first run Main_Inference.r. The initial looping is 50 iterations.
setup.r load group movement reindeer data or simulation data and initialise model parameter and prior
Main_Inference.r the main mcmc loop
RunKF.r run Kalman Filter to calculate the log likelihood
SdeBM.r calculate the diffusion parameter matrix(A in paper) and the initial P for Kalman filter
SolveSdeBM.r calculate the variance matrix (\Xi in paper)
This model can deal with missing values.