cjoint (version 2.1.0)

summary.amce: Summarizing AMCE estimates

Description

summary method for class "amce"

Usage

# S3 method for amce
summary(object, covariate.values=NULL, ...)
# S3 method for summary.amce
print(x, digits=5, ...)

Arguments

object

An object of class "amce", a result of a call to amce.

covariate.values

An optional list containing a vector at which conditional effects will be calculated in the case of AMCE and ACIE's conditional on respondent-varying characteristics. The class of the values in the vector must match the class of the respondent-varying characteristic in question. If the "amce" object contains respondent varying characteristics, when set to NULL (default) interaction effects will be reported at quantiles in the case of a continuous variable and levels in the case of a factor. Names of list entries must correspond to variable names. If there are multiple respondent-varying characteristics then while each is varied in turn, all others will be held at first value of their entry in covariate.values. This is the bottom quantile in the case of a continuous variable and the baseline in the case of a factor variable.

x

An object of class "summary.amce", a result of a call to summary.amce.

digits

The number of significant digits to use when printing.

…

Further arguments from other methods.

Value

The function summary.amce computes and returns formatted data frames of effect estimates returned by amce

amce

A dataframe containing AMCE estimates and standard errors. Each row corresponds to a single attribute-level effect.

baselines_amce

Baseline levels for each attribute relative to which the AMCEs are calculated.

acie

A dataframe containing ACIE estimates and standard errors, if any. Each row corresponds to a single attribute-level effect.

baselines_acie

Baseline levels for each attribute relative to which the ACIEs are calculated.

baselines_amce_resp

Baseline levels for conditional AMCE estimates, if any, relative to which interactions are calculated.

baselines_acie_resp

Baseline levels for conditional ACIE estimates, if any, relative to which interactions are calculated.

samplesize_estimates

The number of valid profiles (rows) in the dataset when only effects of profile varying attributes are calculated.

samplesize_resp

The number of valid profiles (rows) in the dataset when respondent-varying characteristics are incorporated.

numrespondents

The number of respondents in the dataset (if a respondent.id argument was passed to amce)

table_values_amce

A dataframe giving the names of additional tables of AMCE estimates conditional on respondent-varying characteristics. A separate table is produced for each level of each respondent varying characteristic.

table_values_acie

A dataframe giving the names of tables of ACIE estimates conditional on respondent-varying characteristics. A separate table is produced for each level of each respondent varying characteristic.

See Also

The estimation function amce.

Examples

# NOT RUN {
# }
# NOT RUN {
#Results with respondent-varying characteristics
results <-amce(Chosen_Immigrant ~ Gender + Education + Education:ethnocentrism +
	`Country of Origin`+ `Country of Origin`:ethnocentrism + Job +
	Job:ethnocentrism + `Job Experience` + `Job Experience`:ethnocentrism,
	data=immigrationconjoint, design=immigrationdesign,cluster=FALSE,
	respondent.varying="ethnocentrism", na.ignore=TRUE)

#Calculate conditional estimates at user-supplied levels
int.vals<-list()
int.vals[["ethnocentrism"]]<-c(60,77,88,99,45)
summary(results, covariate.values = int.vals)

# }