lybaomc
2009-Oct-17 05:22 UTC
[R] the number of components of the mixture model and the stratfied data
Hi,all with my data,there are more than 1000 quantitative results of antibody concentrations, there may be 2 components(positive and negative), or 3 components (may be strong positive, positive, and negative), or 4-6 components. Could you tell me how to determine the number of components of the mixture model? the "anova.mix" in mixdist of the R software seems not work. my data is a little complicated(TABLE ), there are more than 1000 quantitative results of antibody concentrations stratified by age.The overall density of results at age j, Fj, is a mixture of the component densities, so if there are 5 age groups, then there will be 5 mixture models. Do i have to analyse each stratum respectively? TABLE ------------------------ age Bin length freq ------------------------ 1 1 19.75 4 1 2 21.75 10 ???? 1 12 41.75 36 ???? ------------------------ 2 1 19.75 4 2 2 21.75 10 ???? 2 12 41.75 36 ???? ----------------------- appreciated lybao -- View this message in context: http://www.nabble.com/the-number-of-components-of-the-mixture-model-and-the-stratfied-data-tp25935531p25935531.html Sent from the R help mailing list archive at Nabble.com.
Seemingly Similar Threads
- comparing mixture models (mix function in mixdist package)
- How to use "mix" to estimate the parameters for mixture gamma distribution?
- normal mixture model
- Fitting Mixture of Non-Central Student's t Distributions
- Fitting & evaluating mixture of two Weibull distributions