If climate data were a stock, now would be the time to SELL

Using a financial markets’ trend-analyses tool to assess temporal trend-changes in global surface temperature anomalies (GSTA).

Guest essay by David Dohbro

Heated debates (pun intended) are currently on going regarding if the Earth’s surface temperatures continue to rise, have remained steady, or are decreasing over the past decade or so. To argue for or against any of these three possibilities, pundits often use (linear) regression lines drawn through parts of the different temperature anomaly data-sets that are publically and freely available (GISS, HadCRUT, NCDC, RSS, UAH) to proof or disproof any or all of these possibilities. The problem is that global surface temperatures are none-linear, stochastic in fact; meaning they are dependent on many (random) variables and cycles each operating on many different spatial and temporal scales; natural and possibly man-made alike. Examples are solar activity, volcanic activity, oceanic cycles such as ENSO, PDO, AMO; night/day cycle, seasonal cycle, trace-gasses, cloudiness, etc. Given the nature of the data, the best representation of a temperature trend over time is therefore by using a stochastic time-series trend analyses of the entire data set.

One of the industries where non-linear trend analyses are and have been done over many years is the financial industry. Reason is that asset prices, for example stock and bond prices, are dependent on many variables; are stochastic, and follow non-linear cyclical patterns. In addition, financial markets may often exhibit a directionless trend in time (See Fig. 1; blue horizontal line). However, within such type of larger scale trends smaller scale trends (prices increase and decrease) occur, and financial decisions to either buy, sell or hold assets based on these trends of different time scales need to be made to ensure maximum profits and minimal losses. A rather important task considering we are talking about a daily multi-trillion dollar industry where having accurate and reliable decision tools are obviously paramount.


The Moving Average Convergence-Divergence (MACD) indicator was therefore developed as an additional tool for investors to provide easy-to-interpret (buy and sell) signals, as well as the direction of the price-trend over time[1]. It is a trend-following signal indicator based on three exponential moving averages (EMAs)[2]. The MACD indicator consists of a “MACD Line” and a “Signal Line” (See figure 1; the black and red line, respectively). In this case, the MACD Line is calculated by subtracting the 26-day EMA from the 12-day EMA (See figure 1; the blue and green line, respectively). The Signal Line is the 9-day EMA of the MACD Line. Plotting the MACD Line and Signal Line together with the price data shows how the crossing of these two lines identifies “buy-“ and “sell” signals (See figure 1; the corresponding vertical arrows when the two lines cross), while the direction of the MACD Line identifies the corresponding price-trend. Because the MACD simply subtracts a longer EMA from a shorter EMA it is independent of the nature of the data-set and can be applied to any stochastic (time-series) data set for identification of signals and trends. Theoretically the MACD can thus be applied to global surface temperature anomaly (GSTA) data as well.

Here the MACD is applied to HadCRUT4 data because it is the longest continues data set on record available. First the 12 and 26-year EMAs were calculated from this data, and then subtracted to obtain the MACD. The 9-year EMA was then calculated from the MACD. Both lines were then plotted in the same graph, and the graph placed below the temperature data-set graph on the same time-scale as is done in financial charts (Figure 2). It follows that the MACD of the temperature data peaked or bottomed and then reversed in several instances –see blue vertical lines (Figure 2)- indicating a change of trend in global temperature anomalies; either GSTAs started to increase (~1911, ~1976) or decrease (~1879, 1945, and the latest 2007).

The actual “buy” and “sell” signals (orange arrows) occur a year or two later, because the MACD is a lagging indicator (it is based on longer time-frame moving averages). Note that each and every time these peaks, bottoms and signals occurred in the MACD indicator, temperatures did peak or bottom and subsequently a trend-change occurred: e.g. an increase in GSTA became a decrease and vice versa; no exception. In addition, the MACD also clearly and undeniably identifies the uptrend in temperatures from the mid 1970s till to early 2000s; thought to be the result of mankind’s CO2 emissions; aka anthropogenic global warming (AGW). These “pivot points” validate the yearly-MACD (12, 26, 9) in that it can correctly identify changes in the trends of global surface temperature anomalies reported by HadCRUT4. More about this in detail later.


Now that the MACD-method has been validated we can take a look at the latest signal, which occurred in 2007. The MACD peaked then and has been steadily declining. In addition, the Signal line crossed the MACD in 2008; a “sell” signal occurred. Moreover, the MACD and Signal line are now both pointing down since several years indicating that the temperature trend has changed and the new trend is now down (decrease). Other items of interest that can be deducted from the MACD analyses are the following (See Figure 3):

1) The time-periods between peaks and bottoms in the MACD – blue vertical lines –are of almost identical length (red solid horizontal arrows are of identical length)

2) The increase in MACD (green dotted arrow) is about the same for both periods with increasing GSTA (1911-1945; 1976-2007)

3) The decrease in MACD (yellow dotted arrow) is about the same for both periods with decreasing GSTA (1879-1911; 1945-1976)


What can we learn from these 3 observations? Apparently there are 4 cycles in the current HadCRUT4 data, which suggest GSTAs are now in the next ~32yr cooling period (like any model, we have to work with the data we have and use the past to predict the future). Namely, the MACD of the HadCRUT4 data set finds the following dates with corresponding max and min GSTA values

· max 1879.2 (-0.094), min 1911.7 (-0.362): 32.5yr period

· min 1911.7 (-0.362), max 1945.7 (+0.186): 34.2yr period

· max 1945.7 (+0.186), min 1976.7 (-0.310): 31.0yr period

· min 1976.7 (-0.310), max 2007.0 (+0.829): 30.3yr period

The dates with the actual max and min GSTA values are:

· max 1878.1 (+0.403), min 1911.1 (-0.774): 33.0yr period

· min 1911.1 (-0.774), max 1945.6 (+0.362): 34.5yr period

· max 1945.6 (+0.362), min 1976.2 (-0.439): 30.6yr period

· min 1976.2 (-0.439), max 2007.0 (+0.829): 30.6yr period

The ~32 yr period/cycle; which is an average of these 4 trends becomes apparent, and the MACD does a very good job in determining the dates with the max and min GSTA values. Having determined these dates one can then apply -if one would like to do so- linear regression for each period to determine a slope. Using the actual dates of max, min GSTA values the slopes for each corresponding period/cycle can be determined

· 1879 to 1911: -0.0076°C/yr, R2=0.18 (stat. sign.)

· 1911 to 1945: +0.0141°C/yr, R2=0.52 (stat. sign.)

· 1945 to 1976: -0.0020°C/yr, R2=0.02 (stat. not sign.)

· 1976 to 2007: +0.0193°C/yr, R2=0.64 (stat. sign.)

Using the MACD-determined dates of max, min GSTA-values the slopes for each corresponding period/cycle can be determined

· 1878 to 1911: -0.0066°C/yr, R2=0.15 (stat. sign.)

· 1911 to 1945: +0.0136°C/yr, R2=0.50 (stat. sign.);

· 1945 to 1976: -0.0022°C/yr, R2=0.02 (stat. not sign.)

· 1976 to 2007: +0.0186°C/yr, R2=0.62 (stat. sign.);

It follows, the MACD-determined slopes for each cycle are in very good agreement with those based on using the actual max-, min-GSTA values and dates, showing -again- how accurate and useful the MACD-model is. Point is that stochastic trend and cycle analyses clearly finds periods of about equal length where temperatures rise or decline. The latest cycle, until 2007, indeed saw temperatures rise more rapid, albeit the difference is small, than the previous warming cycle (0.019°C/yr vs 0.014°C/yr; both actual and MACD-determined).

Finally, regression analyses of the data from 2007.0 till 2013.4 shows a slope of -0.002°C/yr and an R2=0.001. Although likely ~25yrs of data for this cooling cycle are still lacking, hence the low R2-value, the slope is already similar to that of the previous cooling cycle. With continuous increasing atmospheric CO2 concentrations since at least 1958 the case can therefore be made that CO2 can not be the main driver in changing GSTA. Instead, the rather similar rates of increases and decreases in GSTAs for the by the MACD identified cycle time-frames, suggest that cycles of around 32 years in length on average, and possibly fractions and multiplications thereof, can explain the observations entirely. The influence of such 30 cycles on Earth’s climate and global temperatures has been reported; e.g. ENSO, AMO, and PDO cycles[3],[4],[5], sea level cycles[6], length of day / atmospheric circulation index cycles[7], solar cycle(s)[8], and planetary cycles[9]. Contrary, these ~32 year cycles are not in sync with global human population/economic activity or to global CO2 concentrations. The latter, instead, increases unabated since 1958[10].

If the current cooling trend is true and applying the ~32yr cycles, it suggests that GSTA should decrease until the late 2030s early 2040s by on average 0.15°C (between 0.06 to 0.24°C) before another warming cycle may commence. Such a cooling trend into the 2030s has been predicted previously[11].

To conclude, this data-analyses tool suggests objectively and without any adjusting, transformation, fitting, “cherry picking” or other means of data manipulation, that GSTA have likely peaked and are now decreasing; a change of trend has occurred. This technique also over comes IPCC’s claim that “Due to natural variability, trends based on short records are very sensitive to the beginning and end dates and do not in general reflect long-term climate trends.” as the more data the better.

[1] Developed by Gerald Appel in the late 1970s. The MACD calculates the difference between two trend-following moving averages; this difference is termed a “momentum oscillator.” The longer period moving average is subtracted from the shorter period moving average to calculate this parameter. As a result, the MACD is an indicator of trend. The MACD fluctuates above and below a zero line as the two individual moving averages converge, cross and diverge over time. See also: http://stockcharts.com/school/doku.php?id=chart_school:technical_indicators:moving_average_conve

[2] Often the 12, 26 and 9-period EMAs are used, where the period can be any suitable time interval from seconds to days to weeks to months and years.

[3] Giese B.S., Ray S. 2011. El Niño variability in simple ocean data assimilation (SODA), 1871–2008. Jounral of Geophysical Research, 116, C02024, doi:10.1029/2010JC006695.

[4] Knudsen et al. 2011. Tracking the Atlantic Multidecadal Oscillation through the last 8,000 years. Nature Communications, 2:178 | DOI: 10.1038/ncomms1186)

[5] www.nwr.noaa.gov/Salmon-Hydropower/Columbia-Snake-Basin/upload/Briefings_3_08.ppt]

[6] Chambers et al. 2012. Is there a 60-year oscillation in global mean sea level? Geophysical Research Letters, 39 (18), DOI: 10.1029/2012GL052885

[7] UN Food and Agricultural Organization (FAO), 2001. Climate Change and Long-Term Fluctuation of Commercial Catches. ftp://ftp.fao.org/docrep/fao/005/y2787e/y2787e01.pdf

[8] http://en.wikipedia.org/wiki/List_of_solar_cycles

[9] Scafetta, N.,2010. Empirical evidence for a celestial origin of the climate oscillations and its implications. Journal of Atmospheric and Solar-Terrestrial Physics, doi:10.1016/j.jastp.2010.04.015.

[10] http://www.esrl.noaa.gov/gmd/ccgg/trends/

[11] Landscheidt, T. New Little Ice Age instead of global warming. Energy and Environment 14, 327-350, 2003.

0 0 votes
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
October 1, 2013 3:31 am

I happened to come across this article by historian David Shearer “Social disorder, mass repression and the NKVD during the 1930s” (link here http://www.jstor.org/discover/10.2307/20174643?uid=3739008&uid=2129&uid=2&uid=70&uid=4&sid=21102715798273 or here http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=2&ved=0CDIQFjAB&url=http%3A%2F%2Fmonderusse.revues.org%2Fpdf%2F99&ei=QY5KUtbzBMKOtQbvx4CwBA&usg=AFQjCNFR9VBreyYlbLKSG7-eosr_xTSFKQ&bvm=bv.53371865,d.Yms).
NKVD, the People’s Commissariat for Internal Affairs was a law enforcement agency of the Soviet Union that directly executed the rule of power of the All Union Communist Party. It was closely associated with the Soviet secret police which at times was part of the agency and is known for its political repression, during the era of Joseph Stalin. While NKVD’s repression was extensive throughout its existence, it really exploded in 1937 and 1938. Is there any particular reason for it?
Shearer writes:
“In late February and early March 1937, several hundred leading functionaries of the ruling Communist Party of the Soviet Union gathered in Moscow for a plenary session of the party’s executive body, the Central Committee. N. I. Ezhov, one of the party’s leading secretaries and head of the People’s Commissariat of the Interior, the NKVD, delivered one of the major speeches at the session. […]
Ezhov’s remarks amounted to a harsh indictment of NKVD policies and a damning criticism of the previous head of the Commissariat, Genrikh Iagoda. Ezhov charged Iagoda and the NKVD with having failed to protect the party and the country from the threat of political sabotage by opposition organizations inside the country and enemy intelligence services working from outside the Soviet Union.”
Iagoda and 3000 of his closest NKVD associates were sentenced to death. Nikolai Ezhov became the new People’s Commissar for Internal Affairs (head of NKVD) and vastly increased its “efficiency”. (It did not really help him in the long run: in August 1938 he was replaced by Lavretiy Beria, in 1939 Ezhov was arrested and in September 1940 sentenced to death.)
The moral of the story is that you are in trouble if you don’t find what your political masters expect you to find. Any similarity with the IPCC is strictly intentional.

October 1, 2013 3:43 am

Short the hell out of NOW!

October 1, 2013 3:48 am

It seems necessary to again point out that ‘global temperature’ is a metric that has no agreed definition and no possibility of calibration. Hence, each Team (GISS, HadCRUT, NCDC, RSS, UAH) that provides a time series of global temperature
(a) uses its own definition of global temperature
(b) changes the definition it uses most months ; see e.g.
Hence, although correct in every way, the above article addresses changes in a meaningless metric.
This is so important that discussion of the matter was suppressed from publication by nefarious method. I again link to explanation of the matter and point to its Annex B for explanation of the problems with the data sets

October 1, 2013 4:01 am

Me thinks September temps AMSU anomaly will be close to 0

Paul Mackey
October 1, 2013 4:18 am

Interesting. It is alos interesting to n ote the markets get it massively wrong too. remember “Long Term Capital management” – more blind faith in models.

October 1, 2013 4:32 am

Hence, although correct in every way, the above article addresses changes in a meaningless metric.

But it’s meaningful to THEM!
I love this approach, especially because I suspect we’re in for a sharp and prolonged decline in global temperatures, because it would be a manifestation of a cosmic come-uppance (nemesis) for hubris. However, as one can see at a glance from the chart, following MACD signals isn’t 100% correct. It can give short-term fake-outs, as it did in 1895 & 1902 (approximately, by eyeball). And it works best in strongly trending markets. In directionless, or “choppy,” markets, the number of short-term fakeouts rises to the point of eliminating the profit–or even delivering a loss for a while.
But trend-following is the only automatic system that “works” in financial markets over the long term. See the four books on turtle trading here: http://www.amazon.com/s/ref=nb_sb_noss_1?url=search-alias%3Dstripbooks&field-keywords=the+way+of+the+turtle
See also the additional books on trend-following by Michael Covel and others, here: http://www.amazon.com/s/ref=nb_sb_ss_i_1_13?url=search-alias%3Dstripbooks&field-keywords=michael+covel+-+trend+following&sprefix=michael+covel%2Cstripbooks%2C250&rh=n%3A283155%2Ck%3Amichael+covel+-+trend+following

Ed Reid
October 1, 2013 4:35 am

At the risk of sounding pedantic, HadCRUT and GISS are not “temperature data sets”. They are temperature anomaly records. The temperatures recorded by the instruments are data. After adjustment, they cease to be data. The anomalies calculated from the adjusted temperature records are also not data.

October 1, 2013 4:38 am

William Delbert Gann, an American investor and a stock market speculator, father of investment charting, one day sitting at the sea shore contemplating natural behaviour of the ocean waves, noticed that an initial small wave is followed by two larger once, each higher then the previous, then the third wave would collapse to the base, for whole sequence to be repeated. He apparently made some ‘serious money’ following this simplest of algorithms.
if this indeed is a natural tendency, the Earth is heading for some serious cooling. No doubt someone may make ‘serious money’ in the process, but millions if not billions around the world will suffer.

October 1, 2013 4:49 am

I came to the same conclusion about future cooling being indicated by looking at the ocean temperature history. Here is what I posted yesterday on another track on this web page .{Bob Tisdale’s letter to John Kerry}
September 30, 2013 at 8:21 am
Well said . The graph that persuaded me of the major impact of oceans on our climate was your Detrended Sea surface temperature Anomalies for the Pacific and Atlantic Oceans Pole to Pole. The peaks in this graph around 1880, 1940 and 2005 and the troughs near 1910 and 1975 match so closely the world global temperature swings . The interesting observations from this graph is that the ocean cycle seems to have peaked and may be heading to a trough by 2045? Cooler weather indicated if ocean SST anomalies are heading down.? This makes a lot more sense to me than a rising co2 which is supposed to raise global temperatures but has not done so for 16.8 years. now. IPCC is projecting temperatures to rise again by 0.2 C per decade . I don’t see this except in isolated El NINO years and even then , there are fewer strong climate altering El Ninos during global cooling cycles[ only one per decade]. The Arctic shows signs of starting to cool , the sun cycle is low and could be so for at least 2 decades more., so there is nothing on the horizon that may raise global temperatures by 0.2 C per decade for the next 20 years . I think IPCC has dug themselves a hole that will be difficult to get out of .

Eddi Rebel
October 1, 2013 4:50 am

In the financial market you follow a sell signal (or buy signal) only because you ASSUME that you have a trend that continues. Often the trend does that. But EVERY SECOND of the market the [trend] can reverse, and it does often, believe me. Often you do not have a trend at all.
Much more can be learned from the financial market if you compare algorithmic trading (trading with an alogrithm) with climate models. I can show you heaps of algos that you can optimize with past data (you take the last 1-2 years for that) and they make heaps of money……in retrospect. They do very [successful] trading…..in the backtest. You go live with that model: it fails miserably. This is called curve-fitting or data mining. It is EXACTLY the same thing climate modelers do – they make a model for something they can’t model properly, fit parameters until they can model the past. They they [run] around and brag about their great model – until it fails in the forward test.

October 1, 2013 4:57 am

At October 1, 2013 at 4:32 am you quote my having said

Hence, although correct in every way, the above article addresses changes in a meaningless metric.

and you reply

But it’s meaningful to THEM!

No! It is USEFUL to them.
And they will choose something else if and when its usefulness ends.
The entire AGW-scare is bogus. It is pure politics hiding behind a mask of pseudoscience.
They predicted warming when they could adjust the surface-based data to indicate warming. Then the satellite data constrained the ‘adjustments’ they can make to the surface-based data, and no warming happened.
Now – e.g. in the above essay – they have got us to play their game by predicting changes to the metrics when
What happens if people with the integrity of Spencer are removed from compiling the satellite data? In that case then I make a prediction; i.e. the satellite data will start to show warming, too.
The AGW-scare was killed at Copenhagen in December 2009. We need to ensure its death throes soon end before its bureaucratic effects are achieved. And to do that we need to kick it into touch, not pass it around the field.

October 1, 2013 5:02 am

Ed Reid:
At October 1, 2013 at 4:35 am you say

At the risk of sounding pedantic, HadCRUT and GISS are not “temperature data sets”. They are temperature anomaly records. The temperatures recorded by the instruments are data. After adjustment, they cease to be data. The anomalies calculated from the adjusted temperature records are also not data.

Your point is not pedantry. It goes to the crux of the entire AGW-scare.
Please see my post at October 1, 2013 at 4:57 am

An Engineer
October 1, 2013 5:03 am

Does the timing of the cross over points of the above lines correspond with the ‘step changes’ proposed by Bob Tisdale in his various graphs?

October 1, 2013 5:09 am

richardscourtney says:
October 1, 2013 at 4:57 am
At October 1, 2013 at 4:32 am you quote my having said

Hence, although correct in every way, the above article addresses changes in a meaningless metric.

and you reply

But it’s meaningful to THEM!

No! It is USEFUL to them.
And they will choose something else if and when its usefulness ends.

But, if they do, they will lose half of their credibility and half their followers–and will make themselves an easy target for relentless mockery. I.e., it will put them on the defensive, which is a bad position for a politician to be in.
(They’re already trying to choose “something else,” with this heat-in-the-ocean dodge. But, besides being something their models didn’t predict (which undermines their credibility), it’s also something that is non-catastrophic, which considerably weakens the case for “action now.”)

kadaka (KD Knoebel)
October 1, 2013 5:11 am

From vukcevic on October 1, 2013 at 4:38 am:

William Delbert Gann, an American investor and a stock market speculator, father of investment charting [cut] (…) He apparently made some ‘serious money’ following this simplest of algorithms.

Wikipedia does not sound kind to WD Gann (6/6/1878 – 6/18/1955):

Gann market forecasting methods are based on geometry, astronomy and astrology, and ancient mathematics.[1][2] Opinions are sharply divided on the value and relevance of his work.[3]

As Harry Houdini showed of the time when he was debunking, there was no shortage of mysticism and superstition among the rich and powerful, who were willing to pay well for “mystical guidance” of many sorts.
The “See also” directs only to the Financial astrology entry. WD Gann is #2 of 2 “Financial astrologers of note”.

October 1, 2013 5:15 am

PS, If the deep oceans can sequester most of the heat, why worry? Problem solved.

kadaka (KD Knoebel)
October 1, 2013 5:16 am

Re previous comment:
I messed up the first link and didn’t catch it in Preview. Here is Wikipedia’s WD Gann entry.

October 1, 2013 5:17 am

At October 1, 2013 at 5:09 am you quote me having said

It is USEFUL to them.
And they will choose something else if and when its usefulness ends.

and you reply by suggesting

But, if they do, they will lose half of their credibility and half their followers–and will make themselves an easy target for relentless mockery. I.e., it will put them on the defensive, which is a bad position for a politician to be in.

History suggests otherwise; e.g. the global cooling was morphed into the global warming scare.
It takes time to ‘convert’ a scare into its successor scare, and that is why the AR5 is trying to ‘buy
time’ when even the contents of the AR5 show ‘the game is up’ for the AGW-scare.
I am arguing that we need to finish off the AGW scare as rapidly as possible for several reasons. One of those reasons is to hinder introduction of its successor scare.

Jeff L
October 1, 2013 5:24 am

David, Nice out of the box analysis ! That is the kind of thinking that ultimately solves problems.
If this “forecast” is correct, it would really reduce climate sensitivity. Assuming the longer ~ linear trend is related to CO2 (an assumption, I know), & we had 0.15 C cooling over the next 25 years while CO2 continues to grow at ~ 2 PPM/yr, my back of the envelope calculation of sensitivity is about ~ 0.9 deg C/ doubling (using pre-industrial times as a starting point). Talk about catastrophe averted ! Even if we continue the “pause” , the data will force climate scientists to reduce their estimates of sensitivity & ultimately admit there is no crisis. We just need time & patience,
The 30 yr cycle that comes out of the analysis is interesting. Joe Bastardi has basically been forecasting this downward trend for many years, first at Accuwx , now at WxBell.

October 1, 2013 5:30 am

A somewhat similar analysis in an article called Predictions of Global Man Temperatures & IPCC Projections by Girma Orssengo was previously posted on WUWT. It was a simple mathematical model or over-fit empirical model based on curve fitting for the GLOBAL YEARLY MEAN TEMPERATURE ANOMALY [GMTA] based on Hadcrut3. The equation or model is not calculated from any measurable parameters other than actual past global temperature anomalies. Although it is not calculated from any physics
* The graph is a general climate trend indicator [up or down]
* There exists a repetitive 60 year climate cycle of 30 years of warming followed by 30 years of cooling.
*There could be two cooling cycles before we reach 2100 which may dwarf and over-ride any greenhouse gas warming
*It is probably more useful and accurate in the short term [next 10-30 years]
*We seem to have peaked on the previous warm cycle and seem to be headed for possible cooling for the next several decades to about 2030

michael hart
October 1, 2013 5:35 am

Hey, if climate data were a stock, the corporate directors would be penniless after multiple class actions by investors, and in prison after criminal prosecutions following SEC investigations.

Doug Huffman
October 1, 2013 5:37 am

Not speaking for him, but from his ‘Anti-fragile’ and ‘The Black Swan’, I believe that successfully retired Quant Nassim Nicholas Taleb would agree. I do wish that he would weigh-in explicitly.

Gary Pearse
October 1, 2013 5:46 am

I’ve long believed that use of these kinds of tools (graced with the moniker “technical analysis”) in markets have actually resulted in self-fulfilling prophecies and are essentially a manipulation of the market. I recall trading commodities many decades ago using moving average prices (14 day?) because that was what I was shown as the method. When the daily price line rose and crossed the MA trace, it was a buy and when it dropped and crossed the MA trace it was a sell. Imagine hundred’s of thousands of traders using this same tool. Triggering a sell, the unanimity of traders indeed did push the price down significantly and vice versa for a buy. Similar things are done with bullion trading. The real supply demand behind legitimate movements was masked, but eventually if serious enough, would assert itself and the pure chart players took a beating. For example, imagine pushing the price up aggressively using the MA signal on orange juice futures at a time when there is a massive frost in Florida. When oranges freeze, they can only be made into orange juice or they are spoiled. Suddenly with aggressive buying, the actual market is soft because of the unexpected jump in supply of orange juice. When the price crosses the MA line on the way down (after traders get the news in the morning paper), it drops like a rock and your sell orders don’t get filled until it is nearly at the bottom. It is the same going the other way on fundamentals.
In 1974, I was in an investment club at work and I had a signal to buy sugar futures at about 10c a lb. Unknown to us a huge sugar shortage was in the making (I can’t remember why – I think it was massive purchases by the Middle East). The decision by the treasurer to wait until 10 o’clock to call it in resulted in our being too late – we forwent a $1100/ cent rise from 10 to ~ 30 to 40 cents because we were shut out by the shear volume of futures purchases – there was a daily limit in contracts and it jumped up limit in minutes for several days.

Doubting Thomas
October 1, 2013 6:06 am

I think the IPCC has sold us a “Norwegian Blue”

Mindert Eiting
October 1, 2013 6:14 am

If I may give the IPCC people an advise, the best way of stopping the present cooling trend is to
re-open the thousands of stations that were closed in the last decade of the former century.

Ed Reid
October 1, 2013 6:20 am

If climate data were a stock, you wouldn’t be able to find the certificates to sell. All you would be able to find would be marked-up fax copies, on which the original number of shares had been “adjusted” in your favor.

October 1, 2013 6:21 am

Pedantic editing point — should none-linear = non-linear?

October 1, 2013 6:25 am

Finally! Something useful on WUWT! I’m going to use it to predict closing prices for stocks today, and get rich. RICH!

gopal panicker
October 1, 2013 6:29 am

HERKHIEMER…is saying what i said 3 years ago

Kevin Cave
October 1, 2013 6:29 am

richardscourtney is correct, and the set-up for the next Big Fat Excuse is already underway.
Since “global air temps” are not behaving in the manner previously portrayed by the warmists, the next meme is already being touted : “ALL THE HEAT IS GOING INTO THE OCEANS”
It is already being injected into the media and I predict will serve the warmists right up to and including the next IPCC report. This one needs to be nipped in the bud right now, in an uncompromising and forthright fashion, before it takes root. “It’s going into the ocean” is their next 4-5 year oxygen, and they need to be deprived of it.

October 1, 2013 6:33 am

Any open minded person can see that the trend has peaked and is now headed back down. Only AGW religious zealots would be blind to this. But nobody has a crystal ball and can predict all the future peaks and valleys.

Don B
October 1, 2013 6:33 am

In 2005 solar physicists Galena Mashnich and Vladimir Bashkirtsev bet climate modeler James Annan $10,000 that global temperatures would be cooler during 2012-2017 compared to 1998-2003. My money is on the Russians.
As a side note, even though Annan was extraordinarily confident at the time of the bet, he has recently been pushing for the climate community to adopt a lower sensitivity number. Kudos to Annan.

gopal panicker
October 1, 2013 6:33 am

predicting stock prices on the basis of these graphs…is a fools game…stick to fundamentals

October 1, 2013 6:41 am

The real question is what is it going to take to get policy makers to stop believing the trash the IPCC is putting out? So far the media (with important exceptions) is still buying it, and without their support there will not be a public outcry that will sway policy makers. Meanwhile, this story is getting lost in the noise over the ACA and the government shutdown.

Fernando (in Brazil)
October 1, 2013 6:45 am

Collapse of Lehman Brothers
Collapse of IPCC 5
Collapse of carbon credits
Calypso as one of the Oceanid daughters of Tethys and Oceanus;
After meeting with Ulysses.[Odysseus]
Hid the heat in the ocean depths.
Aeolus, the master of the winds where he gave Odysseus a leather bag containing all the winds, except the west wind.
Mythology of the XXI century……

October 1, 2013 6:46 am

NH ice back into normal territory!

Silver Ralph
October 1, 2013 6:56 am

Who needs a computer analysis of the data?
Clearly there is a 60-year cycle in the data, which ended in 2010. What else do you need to know??

John W. Garrett
October 1, 2013 6:58 am

Sorry, but any effort to compare trader mumbo-jumbo charlatanry with the concept of investment (see Benjamin Graham and Warren Buffett and the notion of INTRINSIC VALUE) is a non-starter.

RC Saumarez
October 1, 2013 7:00 am

++++++++++ Read McIntyre’s post at climate audit ++++++++++++++++++++
If he is correct, and as far aas I can tell, he is, the IPCC has just engaged in massive fraud.
The answer to the missing heat is simple, just adjust the model outputs so that they appear correct.
I cannot believe the level of fraud and mendacity in the IPPC report. Every UK member of the IPPC should be hauled before Parliament and explain in detail how the IPCC has reached their conclusions.

October 1, 2013 7:06 am

Warren Buffett, a rather successful investor, started his carreer as a charter. he found he could get the same answer when the chart was upside down, and abandonned such foolishness. Multiple billions later he is still outperforming any charter, by using actual data rather than models.

Pamela Gray
October 1, 2013 7:14 am

Not buying the baby with the bath water bit that Richard posits. History repeats itself in terms of the human response to what the climate is doing. Intelligent researchers, armchair enthusiasts, those looking to make a buck, and scary people have worried about the climate and wondered what could be causing a detectable change for hundreds of years. Some try to use it for a power grab, some try to find a reason for the change that lies outside natural variation, some die caught in the occasional violent nature of the change, some look about for the boogyman, some make new insights and understanding is added (eventually), and some just roll with the punch. In this messy response to a warmer weather pattern variation, there has been good science and good discussion on the complexity of the change, regardless of its cause or which data set is better.
A thinking person’s task here is not to throw the baby out with the bath water but to winnow out the bad science from the good, and off-the-cuff conspiracy comments from reasoned responses. It is also a good idea to protect this blog as a place where reasonable discussion is its measure. Without it, we become less than what is desperately needed; a sane voice clearly and loudly spoken. (note to self: stop calling people idiots)
So, with that, in my opinion, the post is a good one. We know that climate naturally swings. This seems a reasonable attempt to describe a model to predict what will happen next.

October 1, 2013 7:19 am

Far be it from me to give investment advice, so take what I say with caution…
From my observation of the stock market and generally successful investors, it seems profitable to go against the trend. So a coarse rule is: when everyone is buying, that is when you should sell, and when everyone is selling, that is the time to buy. Go against the trend when investor sentiment is “too high” or “too low”. So to extend the stock market/IPCC connection, I would say that 97% confidence qualifies as sentiment that is “too high” and now is the time to get out. 🙂

October 1, 2013 7:21 am

I would submit that instead of “four cycles”, Fig.3 actually shows just a bit over 2 cycles.
A cycle is defined as “a series of events that happen repeatedly in the same order”.
Figure 3 shows three peaks and two valleys–hence, just over 2 cycles. Peak to peak would be considered a full “cycle”, as would valley to valley.
Otherwise, a provocative and familiar application since I’ve used stochastic estimates on a number of mining properties. Select inherent confidence limits randomly for each of the contributing factors in a Monte Carlo approach and you can measure overall expectations..
Models can be lotsa fun. But for climate, I’m not convinced we know the confidence limits (sensitivities) of the contributing factors, but obviously CO2 doesn’t fluctuate in 60-year cycles.

October 1, 2013 7:21 am

Correct or not, it is fascinating!
Very well done.
Maybe someone like Wm. Briggs or John Brignell could tell us if they thought all this was Statistically Valid.

October 1, 2013 7:32 am

The author says that global cooling started in 2007.
In 2008 I wrote an article entitled “Is This The Beginning of Global Cooling”, based on UAH data and the announcement by NASA that the PDO had returned to its cool phase.
In retrospect, it was probably too early to make my affirmative conclusion in 2008, and is probably still too early to say so now with confidence.
However, global cooling is looking more probable with each passing year.
Hope I am wrong – humanity suffers during cold periods.
Regards, Allan
(this url no longer works)
Allan MacRae says: February 11, 2012 at 10:42 pm
What has changed since I wrote this article in 2008? Not much, in the big picture. The best fit polynomial isn’t quite as scary, but the Lower Troposphere temperature anomaly is again negative, at -0.09C.
Earth may or may not be cooling just yet, but it sure is not warming anymore.
In fact, Earth has not warmed in over a decade.
What happened to all that “very scary” dangerous manmade global warming? Well, it never really existed. It was all, or almost all natural.
Global warming was, and remains a phony, manufactured crisis – an obvious case of tilting at windmills. History will record global warming as another chapter in “Extraordinary Popular Delusions and the Madness of Crowds”, first published in 1841. Plus ca change, plus ca change pas.
Furthermore, our society just spent a trillion dollars to “fight global warming” – a colossal and shameful waste of scarce global resources.
Who pays the price for this enormous, global warming fraud? We all do.
Wednesday, September 17, 2008
Is This The Beginning of Global Cooling
By Allan MacRae
Many scary stories have been written about the dangers of catastrophic global warming, allegedly due to increased atmospheric concentrations of the greenhouse gas carbon dioxide (CO2) from the combustion of fossil fuels. But is the world really catastrophically warming? NO
And is the warming primarily caused by humans? NO.
Since just January 2007, the world has cooled so much that ALL the global warming over the past three decades has disappeared! This is confirmed by a plot of actual global average temperatures from the best available source, weather satellite data that shows there has been NO net global warming since the satellites were first launched in 1979.
[See image here: http://icecap.us/images/uploads/uah7908.JPG
Since there was global cooling from ~1940 to ~1979, this means there has been no net warming since ~1940, in spite of an ~800% increase in human emissions of carbon dioxide. This indicates that the recent warming trend was natural, and CO2 is an insignificant driver of global warming.
Furthermore, the best fit polynomial shows a strong declining trend. Are we seeing the beginning of a natural cooling cycle? YES.
Further cooling, with upward and downward variability, is expected because the Pacific Decadal Oscillation (PDO) has returned to its cool phase, as announced by NASA this year.
Global warming and cooling have closely followed the phases of the PDO. The most significant pattern of PDO behavior is a shift between “warm” and “cool” phases that last 20 to 30 years. In 1905, the PDO shifted to its “warm” phase. In 1946, the PDO changed to its “cool” phase. In 1977, the PDO returned to its “warm” phase and produced the current warming. In 2007-8, the PDO turned cold again, so we can expect several decades of naturally-caused global cooling.
Some scientists are predicting that this cooling will be severe, and is a greater threat to humanity than global warming ever was. Meanwhile, politicians are still obsessing about global warming.

October 1, 2013 7:33 am

32 years warming. 32 years cooling.
Sounds a lot like the PDO.

October 1, 2013 7:46 am

Has anyone been keeping an eye on Artic Ice trends? It looks like the trend line is zooming right toward the 1979-2012 average.

October 1, 2013 7:54 am

If climate data were a stock, now would be the time to SELL
By David Dohbro

– – – – – – – –
David Dohbro,
Your analysis is a thought stimulator, thanks.
In the situation our modern mixed economy (private voluntary action mixed with gov’t coercive intervention) the signals of a government’s will to and pass history of thwarting private volunteer market activity are also paramount signals to buy and sell. Perhaps, in the case of the current Obama admin in the US, more prudent signals to act on than the actual physical climate signals.
Even initial entry (buying) is intimidated by the rapidly increasing uncertainty of the gov’t interventionist will.

Theo Goodwin
October 1, 2013 7:56 am

richardscourtney says:
October 1, 2013 at 5:02 am
Right. The concept of anomaly was introduced to make trends the basic data of climate science. Nonsense on stilts.

October 1, 2013 7:57 am

I am reminded of the old remark: “If you’re so smart, why ain’t you rich?” The people who use this tool are smart AND rich. It is a stroke of brilliance to apply this purely statistical trend-analysis technique that makes no assumptions to climate data. It also hints at what the primary driver of climate change *really* is.

M Courtney
October 1, 2013 8:10 am

richardscourtney says at October 1, 2013 at 5:17 am

I am arguing that we need to finish off the AGW scare as rapidly as possible for several reasons. One of those reasons is to hinder introduction of its successor scare.

Perhaps, or perhaps not – the next scare is unknown at the moment. It is therefore very hard to evaluate its truth and thus work out if it needs to be opposed.
The next scare may well need opposing as the motivation for creating the scare is not related to any physical reality… but we don’t know what it is, what is wrong with its justification or what the costs it imposes are.
So, perhaps, keeping the present scare alive and slowly humiliating its proponents with a thousand slights and bites out of their funding would be preferable. It allows for both a divide-and-conquer strategy and education of the media as to the nature of scientific scares.
After all we don’t want the flaws in AGW to be forgotten any more than we want them brushed under the carpet. We want to finish the flaws in the media, the scientific community and the political world that allowed this mess to exist in the first place.
PS Happy Birthday to RSCourtney

Alan Robertson
October 1, 2013 8:29 am

The stock market did open this morning and the sun did come up, in spite of the fact the US government is shut down.

October 1, 2013 8:30 am

M Courtney:
Thankyou for your post to me at October 1, 2013 at 8:10 am.
As usual, you and I disagree.
We know the harm being done by distorted economic and energy policies using the AGW-scare as their excuse. You say the next scare may be worse and you could be right, or not.
We have a devil we know, and it needs killing. We can deal with the next devil when we know what it is.

Gail Combs
October 1, 2013 8:51 am

rogerknights says: @ October 1, 2013 at 5:09 am in response to richardscourtney @ October 1, 2013 at 4:57 am
…..But, if they do, they will lose half of their credibility and half their followers–and will make themselves an easy target for relentless mockery. I.e., it will put them on the defensive, which is a bad position for a politician to be in…..
“THEY” do not care because politicians are just bought and paid for tools.

Top Senate Democrat,… Dick Durbin, on a local Chicago radio station this week, blurted out an obvious truth about Congress that, despite being blindingly obvious, is rarely spoken: “And the banks — hard to believe in a time when we’re facing a banking crisis that many of the banks created — are still the most powerful lobby on Capitol Hill. And they frankly own the place.” The blunt acknowledgment that the same banks that caused the financial crisis “own” the U.S. Congress — according to one of that institution’s most powerful members — demonstrates just how extreme this institutional corruption is.
The ownership of the federal government by banks and other large corporations is effectuated in literally countless ways, none more effective than the endless and increasingly sleazy overlap between government and corporate officials…. http://www.salon.com/2009/04/30/ownership/

It is the nameless faceless bureaucrats who actually run governments, politicians come and go but the bureaucrats with their connections to big money actually rule us. The EPA (and the EU and UN) is an example of just how much power these bureaucrats actually have.
To see where the power actually resides you have to follow the ties to the bureaucrats. Just ‘google’ ‘corporate government revolving door’
Then take the next step and look at the research article: The Network of Global Corporate Control by Stefania Vitali, James B. Glattfelder, Stefano Battiston

Tim Ball
October 1, 2013 8:53 am

A good measure that it is time to sell is that some prominent crew members are deserting the sinking IPCC ship. It struck the ice berg of real evidence of cooling.

October 1, 2013 9:01 am

“If climate data were a stock, now would be the time to SELL”
Short sell, in fact

Gail Combs
October 1, 2013 9:11 am

richardscourtney says: @ October 1, 2013 at 8:30 am
…We know the harm being done by distorted economic and energy policies using the AGW-scare as their excuse. You say the next scare may be worse and you could be right, or not.
We also know the next scare has the same motivation. The movement of wealth from the poor and middle class to the wealthy, an increase in power for the bureaucrats and a decrease in freedom for the rest of us.
(Happy Birthday)

October 1, 2013 9:42 am

Doug says:
October 1, 2013 at 7:06 am
Warren Buffett, a rather successful investor, started his carreer as a charter. he found he could get the same answer when the chart was upside down, and abandonned such foolishness. Multiple billions later he is still outperforming any charter, by using actual data rather than models.

Trend-following is not the same as what is meant by old-time “charting”–the Edwards & McGee stuff. Trend-following has had a very successful track record for about 20 years. See M. Covel’s book:

October 1, 2013 9:52 am

No, the time to sell was around COP15 in Copenhagen, late 2009. But if you’re still holding on to the junk, it’s definitely time to GTFO

October 1, 2013 9:59 am

kadaka (KD Knoebel) says:
October 1, 2013 at 5:11 am
Astrology or AGW Cotwology not a great deal of difference.

Freddy F.
October 1, 2013 10:21 am

I find this a great out-side-the box, cross-disciplinary piece of work. We don’t see that very often anymore. Many comments seem to over-analyses this work and take mental leaps (e.g. what has warren buffet to do with all this? he’s an investor not a trader in the first place and he’s always admitted he’s very bad at reading price charts.). It is also great that the author used the entire data set, and not bits and pieces.
This essay’s essence is that the MACD can correctly identify changes in the trends of global surface temperature anomalies reported by HadCRUT4. It does correctly identify the highs and lows almost to the T, which I find striking, and it correctly signals when changes in the trend of GSTA (have) occur(ed). I’ve not seen that being done by any other tool, this simple and this elegant. That’s all there is too it!
This new analytical tool shows that drawing a straight line through GSTAs (from the beginning of each data set) is incorrect, and should be refrained from, since GST and GSTAs are simply non-linear; just like the stock market.
The predictive power of this tool is of course less, and the author could only base any prediction he made on the previous temperature trends of ~32yr down, up, down, up; with each up and down having about the same slope, respectively, which is to me also very striking and not at all in line with the atmospheric CO2 trend. Since one has to work with what one has, that’s as good as it gets. I’d love to see where GSTAs are in 25yrs… IF the MACD finds another low then, we have a winner here!

Doug Proctor
October 1, 2013 10:36 am

All based on the idea that today is not “special”, which is the alarmists bottom line: the past is not a valid indicator of the present, let alone the future.
This is the fundamental point of disagreement in the Climate Wars camps, that the past, including recent obvservations, have relevance to the future. The IPCC use models that do not reflect the past and are not expected to, not even reflect the recent past or present: the “danger” is in the behaviour of the special circumstance of CO2, and lies in the future.
The public doesn’t realise that observation is not important or even relevant when the future is expected to be different from whatever has been before. All history and “common” sense are to be dismissed.

Mike Maguire
October 1, 2013 10:39 am

I trade commodities for a living. Mainly commodities that have weather as a powerful and sometimes dominant driver of price.
Supply side of grains/soybeans is greatly influenced by weather during growing seasons. Demand side of energy markets is greatly influenced by temperatures, especially with respect to residential use of natural gas.
This means my trading decisions are mostly based on the fundamentals of these markets when weather becomes the most important fundamental.
Future weather(forecasts) as the price driver can be identified and used to predict things like crop yields or natural gas use well before the actual weather hits.
The market will dial in the expected changes caused by weather with corresponding price changes. These price changes, in turn cause an effect on the price/market that can be analyzed with many technical indicators, exactly like what the author shows.
In effect, these technical indicators measure the markets reaction to all forces at once being placed upon it. This means that somebody with absolutely no knowledge of any of the forces or fundamentals at work but good at recognizing technical patterns and what they mean, can predict with fair confidence the direction of where price is headed.
There are many traders of stocks and commodities that make a good living and are consistently successful from using only this strategy to predict prices and position for the change.
While I trade fundamentals mostly, I also use and respect all technical analysis. One difference between the atmosphere/climate and stock/commodity prices is that the 2nd one(markets) measures a reaction from people/traders to their expectation of the future while the 1st one is a compilation of a more pure measure using all data.
However, though the 2 of these can diverge in the short run(traders can over react or under react with expectations and price changes) in the long run, markets MUST reconcile to realities and reflect them in price. In other words, an over reaction with price, will be met with selling because of the reality………..an under reaction, with buying when the reality becomes more known.
If instead of actual global temperatures, the measure was the markets expectations of global temperatures, the graph would have much more volatility, with short term spikes up and down based on humans miscalculating and being wrong at times but eventually, the empirical data/measurement would bring it in line with reality.
In both cases, the squiggles and lines on a graph or data on a chart can measure the effect of everything causing them to move added together, allowing an astute technician to interpret the meaning and predict direction without knowing anything at all about fundamentals or what all the individual elements having an effect are.

October 1, 2013 10:45 am

· 1911 to 1945: +0.0136°C/yr, R2=0.50 (stat. sign.);
· 1976 to 2007: +0.0186°C/yr, R2=0.62 (stat. sign.);
That could be used to infer (18.6-13.6)/ 18.6 =27% as being the proportion of the end of 20th c. warming trend that was due to AGW.
After years of parsing data from all kinds of physical situations around the globe my personal IPCC-style “expert judgement” show-of-hands consensus figure is between 25% and 33% .
This estimation technique falls neatly into that range.
Thanks to David Dohbro for this contribution.
Despite the terminology apparently used in finance, these are not running averages, they are a kind of integral. This exponential form is the same as that used to calculate a linear feedback response as is done in so much of climate theory.
Now this just gets me wondering why 9,12 and 26? Why are these the values (in days) used in finance and why do they seem to work in years in climate?
I’ll give this some more thought however, the bottom line of 27% AGW looks very credible to me and the cooling period is similar to that suggested by N. Scafetta and others.

October 1, 2013 11:54 am

This is an very interesting and thought provoking article. I think there are a couple instances where the analogy fails, however. As Mike Maguire points out, there is a human element to stock prices, allowing for the creation and destruction of bubbles, as well as over/undervaluation and emotional trading on smaller scales. Also, as Gary Pearse alluded to, technical analysis in itself can influence prices. Many traders can act upon the same artificial signals, so that the signal itself can disrupt the underlying supply/demand structure given by the fundamentals. With climate, or any other data not affected by human psychology, these aspects of data analyses are absent.
Some other commenters have discussed the validity of using a temperature “data set” for this kind of analysis. I think that point brings up another way in which the stocks/climate analogy does work. The indexes, like the DJIA or S&P 500, are assimilated data with adjustments, with included stocks and weightings varying over time.

October 1, 2013 12:46 pm

Greg says:
October 1, 2013 at 10:45 am
That could be used to infer (18.6-13.6)/ 18.6 =27% as being the proportion of the end of 20th c. warming trend that was due to AGW.

Or perhaps you have identified 27% as being the amount of upward adjustment applied by the Climate Alarmists to the data record.

October 1, 2013 1:54 pm

The MACD analysis is an interesting slant and one which I have been looking forward to reading in this context. As someone who reviews financial forecasts for a living the missed numbers of the IPCC create a natural lack of confidence in what they say. I would not claim that financial market analysis is equivalent to climate models (albeit some of the best mathematical brains in the world “do” technical financial trading). However this type of analysis suggests that other factors are at work. Personally I think that if you look at the picture in the round, the evidence is pointing towards an underlying shift towards cooling. I’ll keep an open mind however, but a close watch on the data – which has proved unreliable and open to manipulation.
The language used will be important too – I am sure we will see more BS about “climate change, climate disruption, dirty weather” etc as temperatures fall…

Gunga Din
October 1, 2013 2:12 pm

Wouldn’t it be kinda’ fun to see Al Gore’s portfolio about now?

October 1, 2013 2:34 pm

I wondered about the relevance of applying stock market analytical tools to Hadcrut data sets but there may be some sense. I have made my own stock market analysis tools and successfully predicted market turning points and index values. The drivers are fundamentally money supply and interest rates. At market bottoms the yield gap between long term government bonds and and the dividend yield on the broad spectrum S&P indexed shares goes to zero. It is possible to model a forward index from the changes in the differences between the two.
Is it possible that the differences in the moving averages are showing some global external physical forces?

October 1, 2013 3:16 pm

Mathematically, MACD is simply an exponential band-pass filter, which reveals the oscillations of signal components near the frequency of peak filter response. It also has interesting phase characteristics, which can lead to turns with no follow-through by the signal. Thus the caveat is to know the spectral structure of the signal quite well. With financial issues that structure is seldom well-known and blind reliance upon MACD has led more than one trader to persistent losses.
We have a similar situation with climate data. HADCRUT4 is a highly manufactured index, whose multidecadal components are not genuine natural signals, but largely the product of PC analysis applied to woefully sparse SST data. The neatly repeatable pattern of ~30yr rises and falls shown here, which looks so good in retrospect, may not hold at all in the future.

October 1, 2013 4:03 pm

“The neatly repeatable pattern of ~30yr rises and falls shown here, which looks so good in retrospect, may not hold at all in the future.”
That is a good point., two similar dips is not enough to establish a reliable cyclic pattern. However, this method does seem to provide a good means of detecting turning points. Willis recently showed 2005 as being the ‘regime change’ based, IIRC, on SST had land+sea. He used a straight cumulative integral rather than exponential integral as used here. The result is nearly identical.
Now even if we cannot count of the next downward segment lasting 32 years, we can use the turning points to determine the start and end points of the warming periods.
Since there was little AGW before WWII, 27% makes 95% sure it’s “more than half” look like a pretty ridiculous claim.

October 1, 2013 4:24 pm

Gary Pearse says:
October 1, 2013 at 5:46 am
“I’ve long believed that use of these kinds of tools (graced with the moniker “technical analysis”) in markets have actually resulted in self-fulfilling prophecies and are essentially a manipulation of the market.”
If many participants start using the same technique, it ALTERS the behaviour of the market until the equilibrium between winners and losers is re-established; it is not a MANIPULATION.

October 1, 2013 4:30 pm

Brian says:
October 1, 2013 at 11:54 am
“Many traders can act upon the same artificial signals, so that the signal itself can disrupt the underlying supply/demand structure given by the fundamentals. ”
The “underlying supply/demand structure given by the fundamentals” is unknown to the market participants; as in practice, no market participant has complete information. Furthermore every market participant extrapolates the fundamentals he believes to know into the future in a different way.
When a majority of traders uses the exact same technique X, technique X automatically loses profitability.

October 1, 2013 4:35 pm

The whole point of my technical comment is that “turning points” in the future cannot be determined reliably by this technique without knowing the (presumably stationary) spectral structure of the true climate signal. I’m far from convinced that anyone has that nailed.

October 1, 2013 6:26 pm

Silver Ralph says:
October 1, 2013 at 6:56 am
Who needs a computer analysis of the data?
Clearly there is a 60-year cycle in the data, which ended in 2010. What else do you need to know??

Please correct me if I’m wrong, but if it ended, it’s not a cycle.

October 1, 2013 7:32 pm

David Dohbro:
Nice article. I get the feeling you are in the neighborhood of chaos theory with some of your references.

Brian H
October 1, 2013 11:08 pm

FerdinandAkin says:
October 1, 2013 at 12:46 pm
Greg says:
October 1, 2013 at 10:45 am
That could be used to infer (18.6-13.6)/ 18.6 =27% as being the proportion of the end of 20th c. warming trend that was due to AGW.
Or perhaps you have identified 27% as being the amount of upward adjustment applied by the Climate Alarmists to the data record.

A serious under-estimate.

Greg Goodman
October 2, 2013 5:52 am

I got to thinking some more about what this “means”.
Economists seem to just do stuff to data and if they like the look of it they keep the result. That goes against my scientific grain but the way this seems to detect the turning points is interesting. That leads me to try to understand what sliding exp means are really doing and why the apparently arbitrary 9.12.26 usually applied in days to stocks , seems to work in years for climate.
Are these “best” values for some reason or is it just luck it works or is the whole thing a coincidence?
Now as I already mentioned convolution with a decaying exponential (which is what these “averages” are really doing) is the way to find the system response to linear negative feedback. And this just happens to be what the whole of climatology seems to be trying to do with climate analysis and modelling.
So the MACD which is the difference between 12 day and 26 day ‘averages’ is, in fact subtracting two different responses to an input signal.
Each linear negative feedback produces a decay response with a different time constant. The MACD is simply the difference assuming both to be of equal size. What this looks like can be seen here:
For example, consider this is the climate system’s response to warming and cooling events, ie the system warms quicker than it cools. Since how Earth captures solar input is very different to how it loses (or hides it in the deep ocean 😉 ) we should not expect this to equal, symmetrical and governed by the same delays.
Now if we excite such a system with a square wave of equal periods (50% mark/space ratio) of a long period like 45 time units, days, years, whatever, Then it will average a bit higher than 50% the input amplitude because it warms quicker than it cools, and it will say around that level.
Now if we shorten it so that the cooling has not quite finished it will start to ramp up. If we back off it will start to cool back to ‘a bit more than 50%”.
Similarly if we vary the mark/space ration we will see similar variations. Now this makes me think of Bob Tisdale’s ENSO ‘driver’. He has suggested for a long time that variations in frequency of Nino/Nina events were the cause of late 20th c. warming.
While I’ve said from the beginning that I think he has made an important observation, my criticism has been that he has found the _mechanism_ not the driver. The next step is to find what is driving variation in ENSO.
While mainstream climatology accepts ENSO has global impact, it regards it as net zero, internal variability. Bob’s point is that this net zero idea is spurious and incorrect.
Now the up-welling of cooler water that characterises La Nina events must be long-term neutral by definition since Nino/Nina events are defined by comparison to long term mean SST.
Whether these deep waves in the thermocline are just random “stochastic” variations of a chaotic system in motion or (as I suspect) some deep water tidal variations, the underlying process can be long term neutral yet changes in frequency or amplitude could cause warming/cooling on the decadal scale. This is Tisdales hypothesis.
David Dohbro’s MACD run seems to tie in with this.
Bring on Willis Eschenbach’s tropical ‘governor’. This suggests that the tropics have a strong built-in capacity to correct for any change in radiative forcing. He does not like the word feedback (which he associates only with linear negative feedbacks ) but a response to change in a system is a feedback of some sort. If it’s self-correcting , it is a negative one. If it has overshoot and restores not just temperature but degree.days then it is a strongly non-linear negative feedback. This is a technically more precise and accurate description than
Now applying Willis’ idea to La Nina we see the tropics will ‘correct’ the cooler SST by adjusting cloud cover and capturing more solar. Conversely, during El Nino, more tropical storms will cut down solar input as well as dumping large quantities of heat into he atmosphere.
Heat dispersed to the atmosphere will eventually radiate to space but there is little chance these opposing mechanisms will have similar response times.
Seeking to understand what MACD is doing and what it may indicate physically in terms of relaxation processes, leads to even totally random ENSO variations causing ‘global warming’ on a decadal or inter decadal scale.
If the ‘pseudo’ cycles of ENSO are in fact a combination of externally driven harmonic cycles as some suggest, we may expect the current cooling to last about 32 years. If it is ‘stochasic’ variation it could be either longer or shorter.
Current indications are that is started somewhere around 2005-2007.

Greg Goodman
October 2, 2013 6:22 am

Recognising that these ‘averages’ are doing something more precise that “smoothing” and adding a fixed delay (in fact they do neither very well) may help to understand why the work of stock data.
The market seems to have a response that can, at least roughly , be thought of as a linear negative feedback. ie as the DOW (or whatever) varies away from the norm investors are going to see reasons to buy or sell , depending on their position. The further away from the norm, the more urgent the need to realign, the greater the motivation to sell / buy.
That describes a negative feedback in the market.
Maybe panic selling is more likely than panic buying to the two processes are asymmetric, as I noted for climate.
The underlying variations are likely stochastic and the asymmetry of market reaction allows MACD to pick up changes. Time constants need to be tuned to the system under analysis. It seems coincidental that what works for markets in days works for climate in years.
I don’t think there is anything ‘magic’ about the numbers other than their ratios are presumably better than some other values. Scaling them depends up on the scale of the events you intend to detect.
Using 26 years leads the method to pick out inter-decadal scale changes.

October 3, 2013 2:26 pm

Greg Goodman:
Exponential bandpass filters, of which MACD is but one example, are straightforward feed-through filters that operate without any feedback loop. Their impulse response is simply a feature of the filter, which tells us nothing about the nature of the input signal or the response characteristics of the system that produced that signal.
I fear that you’re barking up the wrong tree in trying to understand why this approach seems to “work.” If there are fairly narrow-band signal components near the peak frequency response, they will be passed with far less attenuation than those farther away and the indication will look good. In the absence of such strong components, the filter will produce many false “buy/sell” indication, increasingly so as the input approaches gaussian white noise. That is the familiar “whipsawing” in financial markets that kills success. And If you look closely at the so-called “signal line” in Fig. 2, you’ll see that MACD doesn’t work as well as claimed on HADCRUT4. There are misleading “turning points” past the turn of the 20th century and in the 1960s. Caveat emptor!

david dohbro
October 3, 2013 10:05 pm

Thanks everybody for your comments, insights and discussions! I hope I’ve opened up some new ways of thinking and looking at data. Climate science can learn a lot from financial markets where data analyses is of utmost importance given the huge financial implications. Trust me, it has taken me a lot of time to muster the courage to submit this (e.g. note that this data analysis is till 2013.4, whereas there are now two more data points. Nevertheless, the MACD remains below the signal line and both keep decreasing). I noticed that some question the validity of applying a financial trend-analyses tool to climate data (GSTAs), arguing that asset prices are driven –en large- by human emotion, whereas climate data isn’t. However, that is a discussion of the difference between data sets. The MACD is simply an (exponential) moving average based tool. Whatever the type of data, moving averages can be calculated. Now what drives the data is a whole different story. But one first need to know the data trends before one can identify the drivers. The MACD simply identifies the direction of the trend and when a trend changes within the data. Nothing more, nothing less. The article clearly and undeniable shows how well the MACD can identify the low and the high values and trends in GSTAs, objectively. Warren Buffet, investing strategies, once occupation, trading experience, etc have nothing to do with that. Given that the MACD is very accurate in identifying the lows and highs in GSTAs and when the trend in GSTAs changed from increasing to decreasing and vice versa, the fact that since 2008 a decreasing trend in GSTA has been identified is very interesting. What the causes are is a whole other discussion.
I have done the MACD for GISS and NCDC data as well, and both show the exact same MACD pattern as for HadCRUT4. Hence, this isn’t something HadCRUT specific, but GSTA specific. I’ve also done it for NCDC’ northern and southern hemispheric GSTAs and it shows that the MACDs between each hemisphere are much different, with the MACD for the northern hemisphere being much more equal to that of the global STAs. This is also a very interesting fact, and worth a separate discussion. I haven’t done a MACD analyses for RSS and UAH, as their data records are still too short (yes, the MACD wants LOTS of data!).
I wouldn’t go as far as to compare the difference between two slopes and attribute the difference between those then entirely to one cause; in this case AGW. But, assuming that would be the case, then isn’t it amazing that a simple tool like this is able to provide about the same number as what much more complex (and more expensive and time consuming) research and data analyses finds!?
Why the 9, 12, 26 time frames are used (these can be applied to any time frame, from minutes to days to weeks, months, years etc) I am not aware off. It doesn’t make much sense from a market perspective as the week has 5 trading days, but it works and that’s all we need. One could do a sensitivity analyses by using different time frames. Of course the signals become fewer when using longer time frames and more (noisier) when using shorter time frames. Also, the comparison between actual peak and bottom GSTAs with those identified by the MACD shows that the MACD technique is off by only a few months, while using years of time-frames. Hence, it is very sensitive and accurate as it is.
As to the discussion of HadCRUT and GISS are “temperature data sets” or not is semantics. Call these GSTA records what you want, the data analyses remains the same. Bottom line is that it is about finding new ways to analyze the same data to understand the nature of the data better. The MACD is such a new tool, and it clearly identifies periods of warming and cooling that are of similar length and of rather equal rates that are hard to reconcile with constantly increasing atmospheric CO2 levels and exponentially increasing human population and industrialization. It helps underscore the importance of cycles and cycle analyses. IMHO climate science needs to focus on these issues, because everything in this universe goes in cycles.
As for the MACD being a self-fulfilling proficy. That is a fallacy. Ones a buy signal is generated it would in that case mean people will keep on buying and buying and buying for ever. But people don’t. Buying gets exhausted, the buying looses momentum, the MACD starts to point down, and eventually when everybody has bought there’s nothing left buy and only to sell, the MACD will generate a sell signal. Will we then see selling into infinity. No! Similar pattern occurs as with buying. And note that before a buy signal can be generate a sell signal needs to come first, etc… And so the cycles continue; in our market, which are driven by humans, which are driven by natural forces and in nature. It is, IMHO, up to us and climate scientists to identify those cycles. The MACD clearly can help.
Again, thanks all for your comments, etc. Oh and btw, the MACD for the DOW gave a sell signal on the daily yesterday and on the weekly more than a month ago…

October 4, 2013 8:18 pm

Nice. Very nice. I have used Ben Graham fundamentals and MACD both for decades. MACD is very useful for trending data. Many of the complaints about it here are valid for low trend or random signal data. Part of the art is knowing to not use it when trend is weak. As the author says, what it does is spot inflections in trend in data that has trends. Cycles do that. Weather has long duration cycles. It really is that simple. Simple moving average crossovers also do that, but with more time lag.
As to why the same numbers work on day or years, it works with about 100 periods of data to find trend changes in just a few periods. If you want faster or slower response you need to change the numbers used. It doesn’t care what scale is used, just number of periods of data. If your trend inflection takes many more or many fewer time steps to happen, you need to change scale to fit it in 100 to 200 data points. So, for example, use it on 15 minute time steps for daily temp date and it will find sunrise and set with a small lag. Use on 10 second time steps and it will find cloud shadows blowing by in a few minute period.

%d bloggers like this:
Verified by MonsterInsights