Fit LINEX models

Fit LINEX models

linexModel(formula, data, response, x1, x2, x3, time, save.data = TRUE,
  correlation = NULL)

Arguments

formula	a formula of the form response ~ x1 + x2 + x3 + time
data	a data frame in which formula is evaluated
response	instead of specifying a formula, expressions for the components can be specified individually.
x1	instead of specifying a formula, expressions for the components can be specified individually.
x2	instead of specifying a formula, expressions for the components can be specified individually.
x3	instead of specifying a formula, expressions for the components can be specified individually.
time	instead of specifying a formula, expressions for the components can be specified individually.
correlation	an optional nlme::corStruct object describing the within-group correlation structure. See the documentation of nlme::corClasses for a description of the available nlme::corStruct classes. If a grouping variable is to be used, it must be specified in the form argument to the nlme::corStruct constructor. Defaults to NULL, corresponding to uncorrelated errors.

Examples

data(EconUK)
linexModel(iGDP ~ iK + iL + iXp + iYear, data = EconUK)
#> Generalized least squares fit by REML
#>   Model: log(iGDP) - log(iXp) ~ I(2 * (1 - 1/(iK/(0.5 * (iXp + iL))))) +      I(iL/iXp - 1) 
#>   Data: data 
#>   Log-restricted-likelihood: 76.07256
#> 
#> Coefficients:
#>                            (Intercept) I(2 * (1 - 1/(iK/(0.5 * (iXp + iL))))) 
#>                            -0.06059558                             1.12151710 
#>                          I(iL/iXp - 1) 
#>                             2.12109354 
#> 
#> Degrees of freedom: 51 total; 48 residual
#> Residual standard error: 0.04765261 
linexModel(response = "iGDP", x1="iK", x2="iL", x3="iXp", time="iYear", data = EconUK)
#> Generalized least squares fit by REML
#>   Model: log(iGDP) - log(iXp) ~ I(2 * (1 - 1/(iK/(0.5 * (iXp + iL))))) +      I(iL/iXp - 1) 
#>   Data: data 
#>   Log-restricted-likelihood: 76.07256
#> 
#> Coefficients:
#>                            (Intercept) I(2 * (1 - 1/(iK/(0.5 * (iXp + iL))))) 
#>                            -0.06059558                             1.12151710 
#>                          I(iL/iXp - 1) 
#>                             2.12109354 
#> 
#> Degrees of freedom: 51 total; 48 residual
#> Residual standard error: 0.04765261 
naturalCoef(linexModel(response = "iGDP", x1="iK", x2="iL", x3="iXp", time="iYear", data = EconUK))
#>      logscale     scale      a_0      a_1      c_t
#> 1 -0.06059558 0.9412038 1.121517 2.121094 1.891272