Project 28
Project description

In this project, use the models that you set up in Project 23. Because this data is a time series, we should automatically be concerned about correlation between nearby residuals. However, it is not clear the extent to which annual precipitation in a given year is correlated to annual precipitation in nearby years. Investigate the possibility of such correlation in a simple linear regression model of Tacoma precipitation modeled on Seattle precipitation, and in a simple linear regression model of Tacoma precipitation modeled on Portland precipitation. Comment on the extent to which any correlation that you detect should affect the analyses of these models and the inferences drawn from them in Project 23 (where the possibility of such correlation was ignored).

Background on the data set

This data set gives annual precipitation totals for Portland (OR), Tacoma (WA), and Seattle (WA) for the years 1966-2006, precipitation being defined as: "Any form of water particles-liquid or solid-that falls from the atmosphere and reaches the ground."

Data Source: Western Regional Climate Center (, specifically "Monthly Totals" under "Monthly Precipitation Listings" for the following: Portland, Tacoma, and Seattle. The definition of precipitation quoted above is also from this website, specifically in their climate and weather terms glossary. (All links accessed June 14, 2008.)

Variables in the data set
The variables in the data set are as follows:
year(years)year of precipitation measurement
cityPortland, Seattle, or Tacomalocation of measurement
precipitationinchesamount of annual precipiation
Link to the data set
The full data set in csv format is at: