Just in time for the hot weather . . . Aleks points me to this link to a graph of % check-ins at
NYC ice cream shops plotted against temperature in 2011. Aleks writes, “interesting how the ice cream response lags temperature in spring/fall
but during the summer, the response is immediate.”
This graph is a good starting point but I think more could be done, both in the analysis and purely in the graphics. Putting the two lines together like this with a fixed ratio is just too crude a tool. A series of graphs done just right could show a lot, I think!
A scatter plot of checkins versus temperature with the points coloured by a gradient over time (light grey to black)?
[...] –Ice Cream and Temperature: Andrew Gelman points out a fun chart plotting check-ins at ice cream shops against temperature changes. “Just in time for the hot weather . . . Aleks points me to this link to a graph of % check-ins at NYC ice cream shops plotted against temperature in 2011. Aleks writes, “interesting how the ice cream response lags temperature in spring/fall but during the summer, the response is immediate.”” [...]
Checkins seems like a very biased data source. People checking in are making a deliberate statement about their ice cream consumption. I’d be much more interested to see revenue (perhaps just in addition). That would reflect the real impact of temperature.
What’s the definition of percent checkins? I assume it is this:
“Many social networking services, such as Foursquare, Google Latitude, Google+, Facebook, Gowalla, GetGlue and Brightkite allow users to “check in” to a physical place and share their location with their friends.” (from Wikipedia). So this is [checkins involving ice cream shops / total checkins reported].
This is a bit similar to the pattern for farmers’ markets. Crowds lag produce in the the spring, and may stay a bit longer than the produce is available in the fall. That’s a main reason why advertising for farmers’ markets is concentrated early.
reminds me of my favorite Hidden Markov Model tutorial: http://www.cs.jhu.edu/~jason/465/PowerPoint/lect24-hmm.ppt