The problem is separation of multiple normal distributions of fish length, each normal distribution representing a distinct year class contributing to the overall population in a long-lived species. Marian Y Y Yong in 1975 announced a Fortran IV program ENORMSEP in Fishery Bulletin vol. 73 to accomplish this task, combining probit analysis and regression analysis, but did not publish the program. Is there current software out there that can tackle this problem?

  • 2
    $\begingroup$ It sounds to me that you want to fit a mixture of a given fixed number of normal distributions. If so there is a lot of theory for doing this. One approach is maximum likelihood. I expect that there is a lot of software available to do this. However, the ENORMSEP program that you describe sounds like it is a very different approach to this problem. $\endgroup$ – Michael R. Chernick Aug 19 '12 at 13:56
  • $\begingroup$ The VGAM R package may be of help. $\endgroup$ – nico Aug 19 '12 at 14:09
  • $\begingroup$ A quick fit of the distribution can be done in R. stats.stackexchange.com/questions/10062/… $\endgroup$ – bill_080 Aug 19 '12 at 14:42
  • $\begingroup$ In SAS, I think PROC FMM could help with this. $\endgroup$ – Peter Flom - Reinstate Monica Aug 19 '12 at 16:06

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.