Delta method for approximating the standard error of a transformation \(g(X)\) of a random variable \(X = (x_1, x_2, \ldots)\), given estimates of the mean and covariance matrix of \(X\).
Arguments
- g
A formula representing the transformation. The variables must be labelled
x1, x2,...{}For example,~ 1 / (x1 + x2)If the transformation returns a vector, then a list of formulae representing (\(g_1, g_2, \ldots\)) can be provided, for example
list( ~ x1 + x2, ~ x1 / (x1 + x2) )- mean
The estimated mean of \(X\)
- cov
The estimated covariance matrix of \(X\)
- ses
If
TRUE, then the standard errors of \(g_1(X), g_2(X),\ldots\) are returned. Otherwise the covariance matrix of \(g(X)\) is returned.
Value
A vector containing the standard errors of \(g_1(X), g_2(X), \ldots\) or a matrix containing the covariance of \(g(X)\).
Details
The delta method expands a differentiable function of a random variable about its mean, usually with a first-order Taylor approximation, and then takes the variance. For example, an approximation to the covariance matrix of \(g(X)\) is given by
$$ Cov(g(X)) = g'(\mu) Cov(X) [g'(\mu)]^T $$
where \(\mu\) is an estimate of the mean of \(X\). This function
uses symbolic differentiation via deriv.
A limitation of this function is that variables created by the user are not
visible within the formula g. To work around this, it is necessary
to build the formula as a string, using functions such as sprintf,
then to convert the string to a formula using as.formula. See the
example below.
If you can spare the computational time, bootstrapping is a more accurate
method of calculating confidence intervals or standard errors for
transformations of parameters. See boot.msm. Simulation from
the asymptotic distribution of the MLEs (see e.g. Mandel 2013) is also a
convenient alternative.
References
Oehlert, G. W. (1992) A note on the delta method. American Statistician 46(1).
Mandel, M. (2013) Simulation based confidence intervals for functions with complicated derivatives. The American Statistician 67(2):76-81.
Author
C. H. Jackson chris.jackson@mrc-bsu.cam.ac.uk
Examples
## Simple linear regression, E(y) = alpha + beta x
x <- 1:100
y <- rnorm(100, 4*x, 5)
toy.lm <- lm(y ~ x)
estmean <- coef(toy.lm)
estvar <- summary(toy.lm)$cov.unscaled * summary(toy.lm)$sigma^2
## Estimate of (1 / (alphahat + betahat))
1 / (estmean[1] + estmean[2])
#> (Intercept)
#> 0.4072937
## Approximate standard error
deltamethod (~ 1 / (x1 + x2), estmean, estvar)
#> [1] 0.1699689
## We have a variable z we would like to use within the formula.
z <- 1
## deltamethod (~ z / (x1 + x2), estmean, estvar) will not work.
## Instead, build up the formula as a string, and convert to a formula.
form <- sprintf("~ %f / (x1 + x2)", z)
form
#> [1] "~ 1.000000 / (x1 + x2)"
deltamethod(as.formula(form), estmean, estvar)
#> [1] 0.1699689