SP_mlogit: Multinomial logit analysis of a Stated Preference survey

In Iannacks/SP.Survey: SP survey

Multinomial logit analysis of a Stated Preference survey

Description

It provides a multinomial logit summary of a Stated Preference questionnaire made through Google Forms.

Usage

SP_mlogit(attribute_type,dataset_forms,nrespondents,optout,design
base=NULL,conf=FALSE,conf_level=NULL,other_attributes=NULL,sub_sample=NULL)

Arguments

dataset_forms	A character vector containing the file path of each questionnaire made on google form
nrespondents	A numeric vector containing respondents' number for each questionnaire.
design	The experimental design used
attribute_type	A character vector indicating if each attribute must be considered quantitative ("C") or qualitative ("NC")
optout	A logical vector that assumes the value TRUE if the optout option is present, FALSE otherwise.In this case the optout option can be considered a no choice option as well as a status quo option (with always the same attributes levels)
base	A numeric vector indicating the attributes of the no choice or the status quo option
conf	A logical value that assumes the value TRUE if the coefficients have to take the value O if not statistically significant, FALSE otherwise
conf_level	A numeric vector indicating the threshold above which a single coefficient assume the value 0 because not statistically significant
other_attributes	A character vector indicating the variable that you want to use as a filter for the multinomial logit model
sub_sample	A character vector indicating the level of the filtered variable

Details

It provides multinomial logit model estimation.It is derived from a Stated Preference(SP)questionnaire made through Google Forms. One can create a SP experimental design and use the choice tasks found to administer a google form questionnaire. The function takes as input the responses dataset of each questionnaire and the original experimental design from which it creates the final dataset and estimates the multinomial logit model.

Value

multinomial_logit_estimation	The multinomial logit model summary
dataset	The multinomial logit dataset

Author(s)

Gabriele Iannaccone

References

Federov, V.V. (1972). Theory of optimal experiments. Academic Press, New York.

Wheeler, R.E. (2004). AlgDesign. The R project for statistical computing. (http://www.r-project.org).

Croissant, Y. (2012). Estimation of multinomial logit models in R: The mlogit Packages. R package version 0.2-2. URL: http://cran. r-project. org/web/packages/mlogit/vignettes/mlogit. pdf.

See Also

optFederov, optBlock, dcm.design.cand, mlogit

Examples

#design creation
attribute.names=list(delivery_cost = c(3,5,6,10),
                     delivery_time = c("Same day","2/3 days"),
                     delivery_location=c("Pick-up","Home"),
                     co2emissions=c(15,100,150,200,300),
                     carrier_drivers_benefits=c("Low","Medium","High"))
 design<-opt_design(condition=10,alt=2,set=10,block=2,
                    attribute.names=attribute.names,seed=635,
                    sign=c("-","-","+","-","+"))
 design<-design$design
 #Coefficients calculation, don't RUN!!!!
 calculations<-SP_mlogit(attribute_type = c("C","NC","NC","C","NC"),
dataset_forms=c("C:/Users/utente/Downloads/Block 1 DELIVERY (Risposte).xlsx",
        "C:/Users/utente/Downloads/_Block 2 DELIVERY (Risposte).xlsx"),
        nrespondents=c(12,3),design=design,optout=FALSE)
#example with filter
calculations_filter<-SP_mlogit(attribute_type = c("C","NC","NC","C","NC"),
            dataset_forms=c("C:/Users/utente/Downloads/Block 1 DELIVERY (Risposte).xlsx",
            "C:/Users/utente/Downloads/_Block 2 DELIVERY (Risposte).xlsx"),
            nrespondents=c(12,3),design=design,optout=FALSE,
            other_attributes = "Are you a Roma Tre University member?",sub_sample="No")

f2<-SP_market_demand(coefficients=calculations_filter$summary$coefficients,
                     base=c(5,1,1,300,0,0),
                     simulations=data.frame(5,1,1,seq(15,300,by=0.1),0,0),
                     c.var=4,optout=FALSE,design=design,
                     xlab="co2emissions",ylab="market demand",type="l")


f<-SP_market_demand(coefficients=calculations_filter$summary$coefficients,
                    optout=FALSE,simulations=data.frame(5,1,0,100,0,0),
                    base=c(5,1,1,300,0,0),design=design)

k<-partworth_utility(coefficients=calculations_filter$summary$coefficients,
                     design=design,optout=FALSE,
                     attribute_type=c("C","NC","NC","C","NC"))

Iannacks/SP.Survey documentation built on May 18, 2023, 11:23 a.m.