Can use bias corrected confidence intervals by the method of bootstrapping which can be done in r.

see

http://www.google.co.uk/url?sa=t&rct=j&q=&esrc=s&source=web&cd=5&cad=rja&uact=8&ved=0ahUKEwiQuYeRksjNAhULKMAKHbykCXoQFgg-MAQ&url=http%3A%2F%2Fwww.stat.wisc.edu%2F~larget%2Fstat302%2Fchap3.pdf&usg=AFQjCNFirNjV7fJQNvxhn85hIdiQoQVu4g

see

http://www.google.co.uk/url?sa=t&rct=j&q=&esrc=s&source=web&cd=5&cad=rja&uact=8&ved=0ahUKEwjkuaHgksjNAhXsB8AKHcITAfUQFgg-MAQ&url=http%3A%2F%2Fwww.stat.wisc.edu%2F~larget%2Fstat302%2Fchap3.pdf&usg=AFQjCNFirNjV7fJQNvxhn85hIdiQoQVu4g

see

http://www.google.co.uk/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0ahUKEwiR4av3ksjNAhUlL8AKHZ7DBU4QFgghMAA&url=http%3A%2F%2Fwww.statmethods.net%2Fadvstats%2Fbootstrapping.html&usg=AFQjCNFjw5R5JJU16QPn6CvI5gvegbNLrA

download r here

https://www.google.co.uk/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0ahUKEwiyiq-Jk8jNAhViDsAKHRpbAhEQFggcMAA&url=https%3A%2F%2Fcran.r-project.org%2Fbin%2Fwindows%2Fbase%2F&usg=AFQjCNFR55Ylo2A6jQL4cBCL5a9VRJjvJg

Thanks for all your efforts in suggesting answers to my question.  I suspected that some sort of bootstrapping might be needed.  However, I now see that using your proposed methods will require me to learn about r.  That is a good thing to do, but at my age and situation will probably take me about 6 months.  Possibly, I can also leverage samples of similar anthropometric statistics with slightly differing degrees of skewness, asymmetry, and kurtosis to develop a different set of estimation parameters. "Further research is needed" on my part.

Yes, it is possible to estimate them, but possibly their main property would be asymptotic unbiasedness, but not unbiasedness or optimality.

Muchos investigadores y profesionales de las matemáticas y estadísticas, describen los datos de variables no-paramétricas con promedio o media, desviación estándar, error estándar de la media y varianza, lo cual "matemáticamente se puede hacer" pero "No es elegante, tampoco correcto ni adecuado", por otro lado, los datos medidos y descritos así en forma estadística, dan poca validez a los resultados, y por tanto, pone en duda la validez de toda la investigación, desde el diseño, pasando por metodología, instrumentos de medición, hasta resultados y conclusiones; por tanto:

• Si los datos no siguen una distribución normal, como es el caso de las variables cualitativas; nominales, categóricas y ordinales (las cuales pueden seguir una distribución binomial, de Poisson, distribución libre u otra) , se puede utilizar  estadística no paramétrica, la cual puede localizarla en libros, calculadoras científicas y programas estadísticos pre.programados (software) tipo Epistat, Epi-Info, Stata, SPSS, etc.
• En correlación puede utilizar Spearman o Kendall
• Para desviación estándar (correcta sólo en variables paramétricas, verdaderas o continuas) el equivalente en estadística no-paramétrica o cualitativa, es el Espectro, Rango y Recorrido Intercuartílico

Saludos

Dr. José Luis García Vigil

Many researchers and practitioners from mathematics and statistics, describe the data non-parametric variables with average or mean, standard deviation, standard error of the mean and variance, which "mathematically can do "but "not smart, not right nor appropriate ", on the other hand, the measured data and thus described statistically, give little validity to the results, and therefore calls into question the validity of all research, from design through methodology, measuring instruments up results and conclusions; so:

• If the data do not follow a normal distribution, as in the case of qualitative variables; nominal, categorical and ordinal (which can follow a binomial distribution, Poisson, free or other distribution), can be used nonparametric statistics, which can locate in books, scientific calculators and pre.programados statistical programs (software) type Epistat Epi-info, Stata, SPSS, etc.
• In  correlation can be used Spearman or Kendall
• For standard deviation (correct only parametric variables, true or continuous) equivalent non-parametric statistical or qualitative standard deviation is the Spectre, Range and Tour Interquartile.

regards

Dr. Jose Luis Garcia Vigil