Datasets for Statistical Analysis

Chocolates

Keywords:

regression; nutrition; chocolates

Categories:

regression; multivariate analysis
health, popular

Description:

The price, weight and nutritional information was gathered for a number of chocolates commonly available in Queensland stores. The data was gathered in April 2002 in Brisbane.

There are 16 varieties and 7 variables, plus the names of the chocolates are row names.

Variables:

Size	The size of the chocolate (in grams)
Price	The price in dollars
Energy	The kilojoules per 100g
Protein	The percent of protein
Fat	The percent of fat
Carbo	The percent of carbohydrate
Sodium	The mg of sodium

Source:

Collected by a student in one of my classes for Distribution Theory

Notes:

There are different types of chocolates: Can a multivariate technique be used to classify the chocolates into some logical groups (diet chocolates, chocolate bars, candies, etc)? Regression could be used to examine the relationships. Are there some principal components that describe the data well?

References: