Local Warming

Using 55 years of daily average temperatures from a local weather station, I made a least-absolute-deviations regression model that accounts for three effects: seasonal variations, the 11-year solar cycle, and a linear trend. The solution correctly identifies the period and phase of the solar cycle. It also indicates that temperatures have gone up by about 2 ◦F over the 55 years covered by the data. A similar least-squares model failed to characterize the solar cycle correctly. The paper has two goals: (1) to convince climate researchers that median statistics and their generalizations provide a better statistical approach to analyzing climate data and (2) to demonstrate to everyone else how easy it is to analyze at least some climate data for themselves so they can draw their own conclusions.