Invited_talk abstract details

There are many ingredients in stellar models that could in principle be insufficient descriptions of the actual conditions inside of real stars. These deficiencies could easily lead to systematic errors in determinations of the optimal model parameters for a given set of oscillation data. For example, models that employ the relatively simple EFF equation of state lead to estimates of the stellar mass about 10% too high for the Sun, and unacceptably large systematic errors on many of the other stellar properties. Even attempting to ignore the effects of helium settling may be too coarse an approximation, since it leads to 5% errors on the mass. This is not to say that simpler stellar models cannot be used in the analysis of asteroseismic data, but some of the more sophisticated ingredients may be required to obtain accurate results from a global search of the parameter space. I will review these issues with a series of examples using Sun-as-a-star data from the BiSON experiment.