| Type: | Package |
| Title: | Marginal Effects, Odds Ratios and Incidence Rate Ratios for GLMs |
| Version: | 1.2-4 |
| Date: | 2025-05-30 |
| Description: | Estimates probit, logit, Poisson, negative binomial, and beta regression models, returning their marginal effects, odds ratios, or incidence rate ratios as an output. Greene (2008, pp. 780-7) provides a textbook introduction to this topic. |
| License: | GPL-2 | GPL-3 |
| Depends: | stats, sandwich, lmtest, MASS, betareg |
| NeedsCompilation: | no |
| Packaged: | 2025-05-30 09:34:38 UTC; gsl324 |
| Author: | Alan Fernihough [aut, cre], Arne Henningsen [ctb] |
| Maintainer: | Alan Fernihough <alan.fernihough@gmail.com> |
| Repository: | CRAN |
| Date/Publication: | 2025-05-30 10:50:02 UTC |
Marginal effects for a beta regression.
Description
This function estimates a beta regression model and calculates the corresponding marginal effects.
Usage
betamfx(formula, data, atmean = TRUE, robust = FALSE,
clustervar1 = NULL, clustervar2 = NULL,
control = betareg.control(), link.phi = NULL, type = "ML")
Arguments
formula |
an object of class “formula” (or one that can be coerced to that class). |
data |
the data frame containing these data. This argument must be used. |
atmean |
default marginal effects represent the partial effects for the average observation.
If |
robust |
if |
clustervar1 |
a character value naming the first cluster on which to adjust the standard errors. |
clustervar2 |
a character value naming the second cluster on which to adjust the standard errors for two-way clustering. |
control |
a list of control arguments specified via |
link.phi |
as in the |
type |
as in the |
Details
The underlying link function in the mean model (mu) is “logit”. If both robust=TRUE and
!is.null(clustervar1) the function overrides the robust command and computes clustered
standard errors.
Value
mfxest |
a coefficient matrix with columns containing the estimates, associated standard errors, test statistics and p-values. |
fit |
the fitted |
dcvar |
a character vector containing the variable names where the marginal effect refers to the impact of a discrete change on the outcome. For example, a factor variable. |
call |
the matched call. |
References
Francisco Cribari-Neto, Achim Zeileis (2010). Beta Regression in R. Journal of Statistical Software 34(2), 1-24.
Bettina Gruen, Ioannis Kosmidis, Achim Zeileis (2012). Extended Beta Regression in R: Shaken, Stirred, Mixed, and Partitioned. Journal of Statistical Software, 48(11), 1-25.
See Also
Examples
# simulate some data
set.seed(12345)
n = 1000
x = rnorm(n)
# beta outcome
y = rbeta(n, shape1 = plogis(1 + 0.5 * x), shape2 = (abs(0.2*x)))
# use Smithson and Verkuilen correction
y = (y*(n-1)+0.5)/n
data = data.frame(y,x)
betamfx(y~x|x, data=data)
Odds ratios for a beta regression.
Description
This function estimates a beta regression model and calculates the corresponding odds ratios.
Usage
betaor(formula, data, robust = FALSE, clustervar1 = NULL, clustervar2 = NULL,
control = betareg.control(), link.phi = NULL, type = "ML")
Arguments
formula |
an object of class “formula” (or one that can be coerced to that class). |
data |
the data frame containing these data. This argument must be used. |
robust |
if |
clustervar1 |
a character value naming the first cluster on which to adjust the standard errors. |
clustervar2 |
a character value naming the second cluster on which to adjust the standard errors for two-way clustering. |
control |
a list of control arguments specified via |
link.phi |
as in the |
type |
as in the |
Details
The underlying link function in the mean model (mu) is "logit". If both robust=TRUE and
!is.null(clustervar1) the function overrides the robust command and computes clustered
standard errors.
Value
oddsratio |
a coefficient matrix with columns containing the estimates, associated standard errors, test statistics and p-values. |
fit |
the fitted |
call |
the matched call. |
References
Francisco Cribari-Neto, Achim Zeileis (2010). Beta Regression in R. Journal of Statistical Software 34(2), 1-24.
Bettina Gruen, Ioannis Kosmidis, Achim Zeileis (2012). Extended Beta Regression in R: Shaken, Stirred, Mixed, and Partitioned. Journal of Statistical Software, 48(11), 1-25.
See Also
Examples
# simulate some data
set.seed(12345)
n = 1000
x = rnorm(n)
# beta outcome
y = rbeta(n, shape1 = plogis(1 + 0.5 * x), shape2 = (abs(0.2*x)))
# use Smithson and Verkuilen correction
y = (y*(n-1)+0.5)/n
data = data.frame(y,x)
betaor(y~x|x, data=data)
Marginal effects for a logit regression.
Description
This function estimates a binary logistic regression model and calculates the corresponding marginal effects.
Usage
logitmfx(formula, data, atmean = TRUE, robust = FALSE, clustervar1 = NULL,
clustervar2 = NULL, start = NULL, control = list())
Arguments
formula |
an object of class “formula” (or one that can be coerced to that class). |
data |
the data frame containing these data. This argument must be used. |
atmean |
default marginal effects represent the partial effects for the average observation.
If |
robust |
if |
clustervar1 |
a character value naming the first cluster on which to adjust the standard errors. |
clustervar2 |
a character value naming the second cluster on which to adjust the standard errors for two-way clustering. |
start |
starting values for the parameters in the |
control |
see |
Details
If both robust=TRUE and !is.null(clustervar1) the function overrides the robust
command and computes clustered standard errors.
Value
mfxest |
a coefficient matrix with columns containing the estimates, associated standard errors, test statistics and p-values. |
fit |
the fitted |
dcvar |
a character vector containing the variable names where the marginal effect refers to the impact of a discrete change on the outcome. For example, a factor variable. |
call |
the matched call. |
References
William H. Greene (2008). Econometric Analysis (6th ed.). Prentice Hall, N.Y. pp 770-787.
See Also
Examples
# simulate some data
set.seed(12345)
n = 1000
x = rnorm(n)
# binary outcome
y = ifelse(pnorm(1 + 0.5*x + rnorm(n))>0.5, 1, 0)
data = data.frame(y,x)
logitmfx(formula=y~x, data=data)
Odds ratios for a logit regression.
Description
This function estimates a binary logistic regression model and calculates the corresponding odds ratios.
Usage
logitor(formula, data, robust = FALSE, clustervar1 = NULL, clustervar2 = NULL,
start = NULL, control = list())
Arguments
formula |
an object of class “formula” (or one that can be coerced to that class). |
data |
the data frame containing these data. This argument must be used. |
robust |
if |
clustervar1 |
a character value naming the first cluster on which to adjust the standard errors. |
clustervar2 |
a character value naming the second cluster on which to adjust the standard errors for two-way clustering. |
start |
starting values for the parameters in the |
control |
see |
Details
If both robust=TRUE and !is.null(clustervar1) the function overrides the robust
command and computes clustered standard errors.
Value
oddsratio |
a coefficient matrix with columns containing the estimates, associated standard errors, test statistics and p-values. |
fit |
the fitted |
call |
the matched call. |
See Also
Examples
# simulate some data
set.seed(12345)
n = 1000
x = rnorm(n)
# binary outcome
y = ifelse(pnorm(1 + 0.5*x + rnorm(n))>0.5, 1, 0)
data = data.frame(y,x)
logitor(formula=y~x, data=data)
Incidence rate ratios for a negative binomial regression.
Description
This function estimates a negative binomial regression model and calculates the corresponding incidence rate ratios.
Usage
negbinirr(formula, data, robust = FALSE, clustervar1 = NULL,
clustervar2 = NULL, start = NULL, control = glm.control())
Arguments
formula |
an object of class “formula” (or one that can be coerced to that class). |
data |
the data frame containing these data. This argument must be used. |
robust |
if |
clustervar1 |
a character value naming the first cluster on which to adjust the standard errors. |
clustervar2 |
a character value naming the second cluster on which to adjust the standard errors for two-way clustering. |
start |
starting values for the parameters in the |
control |
see |
Details
If both robust=TRUE and !is.null(clustervar1) the function overrides the robust
command and computes clustered standard errors.
Value
irr |
a coefficient matrix with columns containing the estimates, associated standard errors, test statistics and p-values. |
fit |
the fitted |
call |
the matched call. |
See Also
Examples
# simulate some data
set.seed(12345)
n = 1000
x = rnorm(n)
y = rnegbin(n, mu = exp(1 + 0.5 * x), theta = 0.5)
data = data.frame(y,x)
negbinirr(formula=y~x,data=data)
Marginal effects for a negative binomial regression.
Description
This function estimates a negative binomial regression model and calculates the corresponding marginal effects.
Usage
negbinmfx(formula, data, atmean = TRUE, robust = FALSE, clustervar1 = NULL,
clustervar2 = NULL, start = NULL, control = glm.control())
Arguments
formula |
an object of class “formula” (or one that can be coerced to that class). |
data |
the data frame containing these data. This argument must be used. |
atmean |
default marginal effects represent the partial effects for the average observation.
If |
robust |
if |
clustervar1 |
a character value naming the first cluster on which to adjust the standard errors. |
clustervar2 |
a character value naming the second cluster on which to adjust the standard errors for two-way clustering. |
start |
starting values for the parameters in the |
control |
see |
Details
If both robust=TRUE and !is.null(clustervar1) the function overrides the robust
command and computes clustered standard errors.
Value
mfxest |
a coefficient matrix with columns containing the estimates, associated standard errors, test statistics and p-values. |
fit |
the fitted |
dcvar |
a character vector containing the variable names where the marginal effect refers to the impact of a discrete change on the outcome. For example, a factor variable. |
call |
the matched call. |
See Also
Examples
# simulate some data
set.seed(12345)
n = 1000
x = rnorm(n)
y = rnegbin(n, mu = exp(1 + 0.5 * x), theta = 0.5)
data = data.frame(y,x)
negbinmfx(formula=y~x,data=data)
Incidence rate ratios for a Poisson regression.
Description
This function estimates a negative binomial regression model and calculates the corresponding incidence rate ratios.
Usage
poissonirr(formula, data, robust = FALSE, clustervar1 = NULL,
clustervar2 = NULL, start = NULL, control = list())
Arguments
formula |
an object of class “formula” (or one that can be coerced to that class). |
data |
the data frame containing these data. This argument must be used. |
robust |
if |
clustervar1 |
a character value naming the first cluster on which to adjust the standard errors. |
clustervar2 |
a character value naming the second cluster on which to adjust the standard errors for two-way clustering. |
start |
starting values for the parameters in the |
control |
see |
Details
If both robust=TRUE and !is.null(clustervar1) the function overrides the robust
command and computes clustered standard errors.
Value
irr |
a coefficient matrix with columns containing the estimates, associated standard errors, test statistics and p-values. |
fit |
the fitted |
call |
the matched call. |
See Also
Examples
# simulate some data
set.seed(12345)
n = 1000
x = rnorm(n)
y = rnegbin(n, mu = exp(1 + 0.5 * x), theta = 0.5)
data = data.frame(y,x)
poissonirr(formula=y~x,data=data)
Marginal effects for a Poisson regression.
Description
This function estimates a Poisson regression model and calculates the corresponding marginal effects.
Usage
poissonmfx(formula, data, atmean = TRUE, robust = FALSE, clustervar1 = NULL,
clustervar2 = NULL, start = NULL, control = list())
Arguments
formula |
an object of class “formula” (or one that can be coerced to that class). |
data |
the data frame containing these data. This argument must be used. |
atmean |
default marginal effects represent the partial effects for the average observation.
If |
robust |
if |
clustervar1 |
a character value naming the first cluster on which to adjust the standard errors. |
clustervar2 |
a character value naming the second cluster on which to adjust the standard errors for two-way clustering. |
start |
starting values for the parameters in the |
control |
see |
Details
If both robust=TRUE and !is.null(clustervar1) the function overrides the robust
command and computes clustered standard errors.
Value
mfxest |
a coefficient matrix with columns containing the estimates, associated standard errors, test statistics and p-values. |
fit |
the fitted |
dcvar |
a character vector containing the variable names where the marginal effect refers to the impact of a discrete change on the outcome. For example, a factor variable. |
call |
the matched call. |
See Also
Examples
# simulate some data
set.seed(12345)
n = 1000
x = rnorm(n)
y = rnegbin(n, mu = exp(1 + 0.5 * x), theta = 0.5)
data = data.frame(y,x)
poissonmfx(formula=y~x,data=data)
Marginal effects for a probit regression.
Description
This function estimates a probit regression model and calculates the corresponding marginal effects.
Usage
probitmfx(formula, data, atmean = TRUE, robust = FALSE, clustervar1 = NULL,
clustervar2 = NULL, start = NULL, control = list())
Arguments
formula |
an object of class “formula” (or one that can be coerced to that class). |
data |
the data frame containing these data. This argument must be used. |
atmean |
default marginal effects represent the partial effects for the average observation.
If |
robust |
if |
clustervar1 |
a character value naming the first cluster on which to adjust the standard errors. |
clustervar2 |
a character value naming the second cluster on which to adjust the standard errors for two-way clustering. |
start |
starting values for the parameters in the |
control |
see |
Details
If both robust=TRUE and !is.null(clustervar1) the function overrides the robust
command and computes clustered standard errors.
Value
mfxest |
a coefficient matrix with columns containing the estimates, associated standard errors, test statistics and p-values. |
fit |
the fitted |
dcvar |
a character vector containing the variable names where the marginal effect refers to the impact of a discrete change on the outcome. For example, a factor variable. |
call |
the matched call. |
References
William H. Greene (2008). Econometric Analysis (6th ed.). Prentice Hall, N.Y. pp 770-787.
See Also
Examples
# simulate some data
set.seed(12345)
n = 1000
x = rnorm(n)
# binary outcome
y = ifelse(pnorm(1 + 0.5*x + rnorm(n))>0.5, 1, 0)
data = data.frame(y,x)
probitmfx(formula=y~x, data=data)