In: Statistics and Probability
Lurking variables are those that were not included in regression analysis, but are actually the true independent variable. For example, studies show that where there is an increase in ice cream sales, there is also an increase in the number of drownings. Now, ice cream does not affect drowning. The lurking variable in this case is the temperature. As the temperature rises, people buy ice cream and they go swimming. More people swimming will increase the number of drownings. Statisticians need to think about potential lurking variables.
When creating a regression model of what impacts sales at the movies, what independent variables might we include? What might be some lurking variables that would be harder to include directly?