***This blog has been re-launched at: blog.GlowWordBooks.com
What's even better than budgeting? Running reports on financial data you've been collecting over 5 years....
63% - Spent on Groceries
25% - Dining at Restaurants
12% - Dining out, while on vacation
82% - Rainbow Foods (the closest)
16% - Cub (on the way home from most places I go)
1.3% - Lunds (expensive chocolates)
Over 4 years, the amount I pay has trended from $14/fill-up to $17/fill-up.
I have always believed that I would wait to buy gas until I found a good deal... and then buy a lot of gas.... however, running regression analysis on my checkbook shows that the pump price only explains 1% of my purchasing behavior (R^2 = 0.11, F=.706). Days-between-fills is a much more powerful indicator, but still only explains 5% of my purchases (R^2 = .056, F=4.022). Thus, since both variables are very low, I conclude that my gas purchases are fairly inelastic and affected more by personal whims than market changes.
The density of dots in the first third of graph show the massive amounts of driving I did while living in South Dakota (until Jan 04). The the density of dots then shrinks when I moved to Minneapolis and started riding the bus. Finally, the density increases slightly (April 05) when I start driving to uptown to do Comedy Sportz.
South Dakota = 7.5 fills/month
Minneapolis, while riding the bus = 2.75 fills/month
Minneapolis, doing Comedy Sportz = 3.75 fills/month
I couldn't sleep because I know that the price pump has more of an influence on me than just 1% of my purchasing habits, so I began thinking about my methodolgy. Comparing Total Spent and Pump Price is a bad comparison. When the pump price increases, I will buy fewer gallons (lowering my total spent) but it also costs more per gallon (raising my total spent). What I really want to compare is pump price to gallons purchased. When I run that regression analysis, pump price has an R-squared of .078.... which means the pump price explains about 8% of the GALLONS I buy.