Statistical Inference Course Project

The project consists of two parts:

You will create a report to answer the questions. Given the nature of the series, ideally you’ll use knitr to create the reports and convert to a pdf. (I will post a very simple introduction to knitr). However, feel free to use whatever software that you would like to create your pdf.

Each pdf report should be no more than 3 pages with 3 pages of supporting appendix material if needed (code, figures, etcetera).

My Project

This dissertation introduces and empirically validates a new proposal in Electronic Catalogs, based on Visual One Page Catalog (VOPC) [LLW 2003][L&W 2004]. The Theory of Parallel Coordinates is the main theoretical basis for this study. The proposed interface to Electronic Catalogs based on Parallel Coordinates presents, besides filtered products on a single page, multiple attributes in only two dimensions. Among the extensions of original VOPC, it should be highlighted the use of fading and distances between attributes proportional to their relative importance degrees. To evaluate the efficiency of this new electronic catalog, two interfaces have been implemented: traditional (reference for comparison, usually found at portals in the Internet) and Parallel Coordinates. The principal objective of this study is to contribute with the improving of Electronic Catalogs in order to ease product search and choice. Specifically, efficiency (measured by the temporal duration of the search and choice process) and effort (measured by the number of clicks of the mouse) are compared in these two experimental conditions. As result of the (descriptive and inferential) statistical analysis and data interpretation, the equivalence between traditional and Parallel Coordinates was obtained, even being the last one less familiar. Hypothetically, as the proposed system becomes more familiar, favorable significant differences could happen in future experiments.

Key words: Electronic Catalogs, Parallel Coordinates, Internet.

Age Distribution Subject

Dataset

you can load the Dataset, click here

yearsoldsubjects <- read.csv("yearsoldsubjects.txt",header = FALSE)
head(yearsoldsubjects)
##   V1
## 1 50
## 2 50
## 3 50
## 4 50
## 5 50
## 6 50
tail(yearsoldsubjects)
##    V1
## 56 45
## 57 45
## 58 45
## 59 45
## 60 45
## 61 45
colnames(yearsoldsubjects)<-"yo"

The Histogram

h<-hist(yearsoldsubjects$yo,xlab = "Years Old", ylab =  "# Subjects",main = "Age Distribution Subjects",freq = TRUE,col = "GRAY",ylim = c(0,20), xlim = c(30,55),breaks = 8,right = FALSE)
xfit<-seq(min(yearsoldsubjects$yo),max(yearsoldsubjects$yo),length=20) 
yfit<-dnorm(xfit,mean=mean(yearsoldsubjects$yo),sd=sd(yearsoldsubjects$yo)) 
yfit<-yfit*xfit*6
lines(xfit, yfit, col="blue", lwd=2)

Distribution of the experience

This graphicas show the Distribution of the experience in selecttions of accommodations and its attributes on beaches.

Dataset

you can load the Dataset, click here

yearsexpsubjects <- read.csv("yearsexpsubjects.txt",header = FALSE)
head(yearsexpsubjects)
##   V1
## 1  3
## 2  3
## 3  3
## 4  3
## 5  3
## 6  3
tail(yearsexpsubjects)
##    V1
## 59  8
## 60  8
## 61  8
## 62  8
## 63  8
## 64  8
colnames(yearsexpsubjects)<-"yo"

The Histogram

Atention: run install.packages(“htmlTable”), next lines I use this library.

library(htmlTable)
getmode <- function(v) {
   uniqv <- unique(v)
   uniqv[which.max(tabulate(match(v, uniqv)))]
}

h<-hist(yearsexpsubjects$yo,xlab = "# Travels", ylab =  "# Subjects",main = "Travels to Beach",freq = TRUE,col = "GRAY",ylim = c(0,20), xlim = c(2,9),breaks = 20,right = FALSE)
xfit<-seq(min(yearsexpsubjects$yo),max(yearsexpsubjects$yo),length=20) 
yfit<-dnorm(xfit,mean=mean(yearsexpsubjects$yo),sd=sd(yearsexpsubjects$yo)) 
yfit<-yfit*xfit*10
lines(xfit, yfit, col="blue", lwd=1)

idade<-c(mean(yearsoldsubjects$yo), sd(yearsoldsubjects$yo), median(yearsoldsubjects$yo), getmode(yearsoldsubjects$yo), max(yearsoldsubjects$yo), min(yearsoldsubjects$yo))

viagem<-c(mean(yearsexpsubjects$yo), sd(yearsexpsubjects$yo), median(yearsexpsubjects$yo), getmode(yearsexpsubjects$yo), max(yearsexpsubjects$yo), min(yearsexpsubjects$yo))
idade<-round(idade, digits = 2)
viagem<-round(viagem, digits = 2)

myTable <- matrix(c(idade,viagem), ncol=6, byrow = TRUE)

htmlTable(myTable, header = c("mean","sd","median","hiFrenq","max","min"),
          rnames = c("Age", "Travel"),
          caption="Descriptive analysis of the age, number of trips of the subjects",
          tfoot="Table 4.1 from Master Degree Disertation")
Descriptive analysis of the age, number of trips of the subjects
mean sd median hiFrenq max min
Age 44.02 6.31 45 40 55 35
Travel 5.19 1.58 5 4 8 3
Table 4.1 from Master Degree Disertation

The distribution of the buying experience or product attributes of selection accommodation in Brazilian beaches can be seen in Figure 4-3 and Table 4-1 with a descriptive analysis of the collected demographic data (See Appendix III for more references on data demographic of the participants).

The Scientist BR