tau's and sigma's

greenspun.com : LUSENET : CU Statistical Orbit Determination : One Thread

Do we just try different values for each of the tau's in the DMC? Also for eq(28) in the DMC notes on the web, is sigma^2 a scalar? Do you have any suggestions as to what values to use?

-- Rodney Anderson (rla@colorado.edu), February 18, 2000


Start with values for tau around 1e-7 and increase by a factor of 10 until it changes the results. Since you are doing RIC you need tauR, tauI, tauC, sigmaR, sigmaI and sigmaC. The handout I gave you is better than what is on the web. For those doing xyz, you really only need one tau and sigma.

-- David Goldstein (david.goldstein@colorado.edu), February 18, 2000.

Moderation questions? read the FAQ