suppressPackageStartupMessages(library("foehnix"))

Foehnix Demo

foehnix objects also provide asymptotic inference of the estimated coefficients of the mixture model (see statistical model and logistic regression with IWLS).

Let's start with estimating our demo flexmix foehn diagnosis model. We have already prepared the data sets (object data). Details how to generate the data object and more information about the foehnix model specification can be found on the getting started manual page.

# Loading combined demo data set
data <- demodata("tyrol") # default
# Estimate the model
mod <- foehnix(diff_t ~ ff + rh, data = data)

The summary method prints the test statistics for both parts of the mixture model, the component model and the concomitant model (if specified).

summary(mod, detailed = TRUE)

Component Model Inference

The inference for the two location parameters $\mu_1$ and $\mu_2$ of the two components are based on the asymptotic theory. We expect that the estimates of our coefficients are unbiased. Thus, the expectation of our estimated coefficients is the estimated coefficient itself ($\text{E}(\hat{\mu}_1) = \mu_1$, $\text{E}(\hat{\mu}_2) = \mu_2$).

In a general form the covariance matrix of a liner model (one component, unweighted) for a set of $i = 1, \dots, N$ observations can be expressed as follows:

... where $N$ is the sample size, $P$ the number of parameters or covariates, $\mathit{\epsilon} = \mathit{y} - \mu_\bullet$ the model residuals, and $\mathbf{X}$ the model matrix of the linear model. In case of a foehnix model we have two components where each component consists of an intercept only model ($\mu_1$ and $\mu_2$ do not depend on additional covariates). Thus, $P = 1$ and $\mathbf{X}$ is an $N \times 1$ matrix with 1s. The estimates are based on a set of weighted observation $y$, in this example diff_t ($y =$ diff_t). The weights are the a-posteriori probabilities $\hat{\mathit{p}}$ (the foehn probabilities) of the foehnix model and have to be taken into account when calculating the standard errors. With these weights the standard error for component 2 can be written as:

... or much simpler:

The same holds for component 1 except that our weights are $1 - \hat{\mathit{p}}$:

Concomitant Model Inference

If a concomitant model has been specified summary will also return the corresponding z statistics for the estimated regression coefficients of the logistic regression model (see logistic regression with IWLS).

The covariance matrix of a logistic logistic regression model with the regression coefficients $\mathit{\alpha}$ (with $P$ coefficients) with a dispersion parameter of 1 of the binomial family is given as:

where $\mathbf{X}$ is the model matrix of the concomitant model (un-standardized) of dimension $N \times P$ and $\mathit{\omega} = (\mathit{\pi} * (1 - \mathit{\pi}))^\frac{1}{2}$. $\pi$ is final response, probability returned by the logistic regression model. The diagonal of the covariance matrix contains the variances of the estimated regression coefficients. Thus,



retostauffer/Rfoehnix documentation built on June 5, 2023, 11:39 p.m.