---
title: "Estimating the exponential of regression parameters using **brglm2**"
author: "[Ioannis Kosmidis](https://www.ikosmidis.com)"
date: "03 February 2023"
output: rmarkdown::html_vignette
bibliography: brglm2.bib
vignette: >
  %\VignetteIndexEntry{Estimating the exponential of regression parameters using **brglm2**}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r setup, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.width = 6,
  fig.height = 6
)
```

# The `expo()` method

The [**brglm2**](https://github.com/ikosmidis/brglm2) R package provides the `expo()` method for estimating exponentiated parameters of generalized linear models using various methods. 

The `expo()` method uses a supplied `"brglmFit"` or `"glm"` object to estimate exponentiated parameters of generalized linear models with maximum likelihood or various mean and median bias reduction methods. `expo()` is useful for computing (corrected) estimates of the multiplicative impact of a unit increase on a covariate on the mean of a Poisson log-linear model (`family = poisson("log")` in `glm()`) while adjusting for other covariates, the odds ratio associated with a unit increase on a covariate in a logistic regression model (`family = binomial("logit")` in `glm()`) while adjusting for other covariates, the relative risk associated with a unit increase on a covariate in a relative risk regression model (`family = binomial("log")` in `glm()`) while adjusting for other covariates, among others.

The vignette demonstrates the use of `expo()` and the associated methods by reproducing part of the analyses in @agresti:02[, Section 5.4.2] on the effects of AZT in slowing the development of AIDS symptoms.

# AIDS and AZT use

The data analyzed in @agresti:02[, Section 5.4.2] is from a 3-year
study on the effects of AZT in slowing the development of AIDS
symptoms. 338 veterans whose immune systems were beginning to falter
after infection with the AIDS virus were randomly assigned either to
receive AZT immediately or to wait until their T cells showed severe
immune weakness. See `?aids` for more details. 

The `aids` data set cross-classifies the veterans' race (`race`), whether they received AZT immediately (`AZT`), and whether they developed AIDS symptoms during the 3-year study (`symptomatic` and `asymptomatic`).
```{r, echo = TRUE}
library("brglm2")
data("aids", package = "brglm2")
aids
```

We now use a logistic regression model to model the probability of developing symptoms in terms of `AZT` and `race`, and reproduce part of the compute output in @agresti:02[, Table 5.6].
```{r, echo = TRUE}
aids_mod <- glm(cbind(symptomatic, asymptomatic) ~ AZT + race, 
                  family = binomial(), data = aids)
summary(aids_mod)
```

The Wald test for the hypothesis of conditional independence of AZT treatment and development of AIDS symptoms, controlling for race, returns a p-value of `r round(coef(summary(aids_mod))["AZTYes", "Pr(>|z|)"], 3)`, showing evidence of association.

The predicted probabilities for each combination of levels


The maximum likelihood estimates of the odds ratio between immediate AZT use and development of AIDS symptoms can be inferred from `aids_mod` through the `expo()` method, which also estimates standard errors using the delta method, and returns approximate 95% confidence intervals (see `?expo` for details).
```{r, echo = TRUE}
expo(aids_mod, type = "ML")
```

As noted in @agresti:02[, Section 5.4.2], for each race, the estimated
odds of symptoms are half as high for those who took AZT immediately,
with value $0.49$ and a nominally 95\% Wald confidence interval
$(0.28, 0.84)$.

The `expo()` method can be used to estimate the odds ratios using three methods that return estimates of the odds ratios with asymptotically smaller mean bias than the maximum likelihood estimator 
```{r, echo = TRUE}
expo(aids_mod, type = "correction*")
expo(aids_mod, type = "Lylesetal2012")
expo(aids_mod, type = "correction+")
```
and one method that returns estimates of the odds ratios with asymptotically smaller median bias than the maximum likelihood estimator 
```{r, echo = TRUE}
expo(aids_mod, type = "AS_median")
```

The estimated odds ratios and associated inferences from the methods that correct for mean and median bias are similar to those from maximum likelihood.

# Infinite odds ratio estimates 

When `expo()` is called with `type = correction*`, `type = correction+`, `type = Lylesetal2012`, and `type = AS_median`, then the estimates of the odds ratios can be shown to be always finite and greater than zero. The reason is that the corresponding odds-ratio estimators depend on regression parameter estimates that are finite even if the maximum likelihood estimates are infinite. See, @kosmidis:2019 and @kosmidis+firth:21 for details. 

As an example, consider the estimated odds ratios from a logistic regression model fitted on the `endometrial` data set using maximum likelihood. 
```{r, echo = TRUE}
data("endometrial", package = "brglm2")
endometrialML <- glm(HG ~ NV + PI + EH, data = endometrial, family = binomial())
endometrialML
```

The estimate of the coefficient for `NV` is in reality infinite as it can be verified using the [**detectseparation**](https://cran.r-project.org/package=detectseparation) R package
```{r, echo = TRUE}
library("detectseparation")
update(endometrialML, method = detect_separation)
```
and a naive estimate of the associated odds ratio while controlling for `PI` and `EH` is `r exp(coef(endometrialML)["NV"])`, which is in reality infinite.

In contrast, `expo()` returns finite reduced-mean-bias estimates of the odds ratios 
```{r, echo = TRUE}
expo(endometrialML, type = "correction*")
expo(endometrialML, type = "correction+")
expo(endometrialML, type = "Lylesetal2012")
```

# `brglmFit` objects

The `expo()` method also works seamlessly with `brglmFit` objects, returning the same results as above. For example, 
```{r, echo = TRUE}
aids_mod_br <- update(aids_mod, method = "brglmFit")
expo(aids_mod_br, type = "correction*")
```


# References