hSkew function

Hougaard Measure of Skewness

Hougaard measure of skewness with nonlinear regression

hSkew(rx)

Arguments

• rx: a result of nls function

Details

Hougaard measure of skewness can be used to check if the parameters of nonlinear regression behavior in linear fashion, i.e. symmetric confidence interval. Be cautious on the variable name conflict. All the variables in the nonlinear function should be able to be accessed by the function.

Returns

Hougaard estimate of skewness for each parameter - (0, 0.1]: The estimate is very close-to-linear in behavior.

• (0.1, 0.25]: The estimate is reasonably close-to-linear in behavior.

• (0.25, 1]: The skweness is apparent.

• >1: The estimate is considerably nonlinear in behavior.

Examples

r1 = nls(density ~ b1*conc/(b2 + conc), DNase[DNase$Run == 1, ], start=list(b1=3, b2=5))
  hSkew(r1)

References

EL-Shehawy SA. On Calculating the Hougaard Measure of Skewness in a Nonlinear Regression Model with Two Parameters. J Math & Stat. 2009;5(4):360-364.

Author(s)

Kyun-Seop Bae k@acr.kr