Seasonality in Painting Production?

Analytics.

Josiah Stamp once remarked that,

"The government are very keen on amassing statistics. They collect them, add them, raise them to the nth power, take the cube root and prepare wonderful diagrams. But you must never forget that every one of these figures comes in the first instance from the chowky dar (village watchman in India), who just puts down what he damn pleases."

Well! Within my painting logs are amassed years of data none of which follow Stamp's Law.  You see, I only add to the ledger figures actually completed.  As I look over the record, I often wonder if there is any rhyme or reason to my painting output.  Looking at production since 1996 on an adjusted painting points basis (1), not many patterns are discernible at this aggregated level.

Monthly Painting totals by Month and Era 

Generally, it appears that both output and number of eras have increased over time but not much more can be identified.  I typically breakdown these numbers into an annual review offering more detail but what about a seasonal nature to my output?  Anecdotally, my memory tends to latch onto the notion that painting increases during the short daylight hours of winter and slacks off during the long and warm days of summer.  Do these data support that conclusion?  Let us put it to the test.

First, begin by overlaying a trend onto the monthly painting points totals (TOTPOINT). Yep.  In general, painting production has increased over time.  With seemingly periodic crests, perhaps, there is some seasonality within the data?

Monthly Painting Totals with Trend 

Using the 19 years of history and decomposing the time series shows that a definite seasonal pattern does emerge.  What does the graph below suggest?  The analysis suggests that after seeing an increase in production in February, production drops during the spring and summer months with output roughly 50 adjusted painting points below the monthly mean output.  Then in August, I tend to hit the paint desk with vigor only to drop off again in September.  After the September lull, October witnesses another spike in activity that carries on through the remainder of the year. 

Seasonal Patterns in Painting Production

Interesting?  Well, this exercise provides insight into my painting habits and provides reasonable inferences.  The relative spikes in FEB, AUG, and late fall also coincide with my experiences.  Did not realize my habits were so predictable.  I guess at least I am  a creature of habit.

Notice that each month with an increase in activity is followed by a month with decreasing activity.  In statistical terminology, the series exhibits negative serial correlation.  That is, an increase in one month is followed by a decrease in the next month.  This pattern suggests that painting produces fits and spurts where increased productivity in one month leads to, perhaps, burnout in the following month.  Yeah, I have felt that on occasion.  The data suggest it happens every other month!

With the evidence suggesting that I am, indeed, a creature of habit, tomorrow I celebrate Christmas!

Merry Christmas to all!

(1) Adjusted Painting Points based on Analog Hobbies points scoring system. Adjusted Paintings Points tend to weight all work appropriate to effort rather than based solely on raw totals of painted figures.