Regressions - mattwigway/bikeshare-analysis GitHub Wiki

We have lots of variables to choose from. Interesting is that neither housing nor jobs within 60 minutes give a very good R^2 (< 0.02), but together they give adjust R^2 of 14%:

Coefficients:
            Estimate Std. Error t value Pr(>|t|)    
(Intercept)  95.5215    16.0741   5.943 1.04e-08 ***
housing60     3.7252     0.6112   6.094 4.61e-09 ***
jobs60       -2.9667     0.5200  -5.705 3.58e-08 ***

What's interesting is that the coefficient on jobs is negative. This is counter-intuitive, but perhaps there is an effect of density: is it possible that, at very high densities, bikesharing is inconvenient?