Posts filed under “Quantitative”
"Not only do we misunderstand Science, we’re bad at math, too!"
So say the innumerates.
Every now and then, I venture over to other fields to see what the
debates look like. The most recent laugher was amongst the global
warming denialist crowd.
Why? In 2007, the average global temperatures dropped by 0.595 degrees
centigrade. This is a fact. The response from this group was to say (verbatim) "Twelve-month long drop in world temperatures wipes out a century of warming."
Um, no. As the charts below reveal, it does nothing of the sort.
As we say all the time with the Non-Farm Payroll (NFP), you
look at the overall trend, not any single data point. A monthly NFP of 10,000
does not guarantee a recession, nor does a single month of 200k
job gains guarantee an expansion. And neither monthly release eliminates the trend of
the prior 100 data points.
All data series have anomalies — large magnitude points that may be curious, or unusual. To claim that a "Twelve-month long drop in world temperatures wipes out
a century of warming" simply reveals a disturbing
statistical/mathematical incompetency that is rather embarrassing.
The 20 year and 130 year charts clearly explain what this quite clearly.
The shorter term chart shows a volatile series, with high magnitude aberrations to the upside (1998) as well as the downside (2008):
The longer term chart unequivocally reveal a long term trend, as the data points move from the lower left to the upper right
Boo! Global Warming Remains an Ongoing Trend (1880-2010)
Sources: Watts Up With That?
If the above long term chart was a stock, would you short it?
Now, my own views on Global Warming are spectacularly conventional: Take 6 billion people, give them an industrial revolution; then for the next 2 centuries, have them burn all manner of carbon products. Its not too hard to imagine this activity might impact the system in which it takes place.
These aren’t the ravings of an enviro-nut. I am a big time energy consumer. I don’t lecture anyone about their energy consumption. Anytime someone offers me a ride on a private jet, I jump at it. Yes, I dislike SUVs — but that’s because they are ungainly, unsafe, handle poorly, and go-too-slow. I personally have way too many motorized vehicles, all but one of which skew towards high-horsepower, go-fast, poor-fuel economy end of the scale. Trust me when I tell you a 6 speed manual transmissions in a V12 is not about saving gasoline.
Another disclosure: I have been long oil for 5 years, and quite a few oil companies for much longer. I am not a shrinking violet when it comes to recognizing the ongoing demand for energy, and the role that carbon based products are likely to play over the next decade or longer. I have personally profited hugely from these oil positions.
But I am at heart someone who loves math and statistics, and who finds the abuse of the truth to be offensive. Anyone who claims that a high magnitude outlier within a volatile data series conclusively proves this or that — someone who chooses to ignore the broader data trend — is simply putting their own mathematical ignorance and innumeracy on display.
Your mileage may vary . . .
UPDATE: March 1, 2008: 1:45PM
Do not misunderstand my position: I am not advocating pro or con for any specific policy, nor am I arguing against nuclear power.
What I said above is the person who made the statement that the past 12 months average temperature decline has wiped out a century of global warming is a clueless innumerate.
I appreciate the many intelligent statements in comments. My beef is with the chart and the math, not the policy discussion.
4 sources say “globally cooler” in the past 12 months
Watts Up With That? 19 February 2008
The Innumerate (mathematically illiterate):
Temperature Monitors Report Widescale Global Cooling
Daily Tech February 26, 2008 12:55 PM
Evidence of Global Cooling
Fox News, Thursday, February 28, 2008
Apple (AAPL), Google (GOOG) and Baidu (BIDU) — three of the NASDAQ FOUR HORSEMAN – have been coming up gimpy since their FusionIQ timing sell signals (triggered at much higher levels several weeks ago).
These stocks have fallen precipitously since those Sells.
Only Research in Motion (RIMM) — which gapped open strongly yesterday — acts well. The other three horsemen stocks have clearly broken their uptrends, and are in the process of being repriced. We point this out because trading highly volatile names like these can be dangerous especially if one doesn’t have a Sell Discipline.
Unbiased, indicator-driven trading signals can help make you a better investor. Whether its based on technicals, fundamentals, valuation or quantitative research, having a non-emotional layer to your investing/trading plan is always helpful.
I am biased towards our signals (FusionIQ), but any objective timing method that would have gotten you out of the way in these 3 names weeks ago is a good thing — you would have avoided a lot of pain.
Four Horseman of Nasdaq
There are lots of things that investors believe which I find perplexing. The Superbowl indicator is one, but the oddest to me is the so-called Fed Model, also known as the IBES Valuation Model. It is not that the Fed model is so terribly wrong — it has been both right and wrong over the…Read More