This preview shows page 1. Sign up to view the full content.
Unformatted text preview: bles • Often categorical variables are coded with numbers • Look like numerical variables but really are NOT – really important! 16 NUMERICAL VARIABLES Numerical variables can be either discrete or continuous or approximately continuous Definition A discrete variable is one where we can count the number of possible values of the variable. Examples: X = number of children X = number of new patents (sample of families) (sample of annual data for chemical companies) Other discrete variables? 17 NUMERICAL VARIABLES Definition A continuous variable is one that can take on any value in a continuum—any value on the real line, or in an interval of the real line. Examples X = wind speed (sample of daily data for several locations near wind farms) X = hours to assemble/finish a vehicle (sample of vehicles from assembly lines) X = emissions (sample of monthly data for factories) Other continuous variables Probability models for discrete and continuous variables are fundamentally different. 18 NUMERICAL VARIABLES A discrete variable can be treated (modeled) as continuous, and is said to be approximately continuous, if the following conditions are met: 1) the unit of measurement is small compared to the typical value of variable (or the range of values of the variable) 2) the number o...
View Full Document
- Spring '08