‘Lies, damned lies, and statistics’: Forecasters are never wrong - Even when they are wrong.

Ok so this is not about trading or investing, but how easily could we apply the following to the many market forecasters, predictors, gurus and scribblers.

The following is taken from an article on the BBC website in relation to the appalling medium-term/long-range forecasting record of the British Meteorological Office (The Met Office). Full article can be seen here:  http://www.bbc.co.uk/news/science-environment-21967190.

To summarise; last spring the UK suffered a switch from extremely dry almost drought-like conditions to incredibly wet with heavy downpours and seriously flooding. The Met office had stopped publishing long-range weather forecasts after a series of terrible high-profile failures in recent years; however they continued to publish ‘secret’ advisories to government contingency planners. – The article relates to the prediction for spring last year, issued in March. The 3 month advisory stated:

•  "The forecast for average UK rainfall slightly favours drier than average conditions for April-May-June, and slightly favours April being the driest of the three months.”  

Fast forward over the next 3 months, and these were examples of newspaper headlines:

• ‘Wettest April in the UK for over 100 years’
•  ‘Areas of UK see three times their average rainfall in spring’
•  ‘UK environment agency: Three months April to June - More rainfall than at any time since records begun.’ (Which was in 1910).

A failing that is so completely obvious, that to even try to deny it would just be incredulous. – And yet, full marks to the Met Office’s chief scientist Prof Julia Slingo who says:

•  ‘Last year’s calculations were not actually wrong because they were probabilistic’.

Hhhhmmmmmm. Let’s just repeat the first part of that quote:  ‘Last year’s calculations were not actually wrong’. 

  – It remains me of a wonderful quote by Charles Darwin: ‘A Mathematician is, a blind man, in a dark room, looking for a black cat, which isn't there”.

It might just be more sensible for them to admit that Yogi Berra is right: 

Finally - In keeping with the theme. - Here is a fantastic video clip called 'African Rain'. Try listening to it first without watching. - Then watch it.

'Rain_Rain_Rain' image courtesy of Christian Southworth / FreeDigitalPhotos.net.

Some further thoughts on forecasting?

I recently posted a couple of articles about forecasting and whether there is any utility achieved from forecasting. - The first of these was posted a few weeks a ago and can be seen at this link 'Is it folly to forecast?' this was followed up by a question on my 'LinkedIn group' - 'Is forecasting a waste of time?' - which was intended to stimulate debate rather than express my opinion, and which certainly produced some interesting responses.

Before, I proceed any further I also want to refer to an article by Michel Pireu in 'Business Day' from last year where he talks about James Montier's view on forecasting - the article can be seen here. I have discussed some of Montier's views previously, I am a huge fan of Montier's work, not only does he make some excellent points and observations, his writing style is both interesting and uncomplicated, which means unlike so many books on aspects of trading and finance, you won't need matchsticks wedged into your eyelids to stop them from shutting. - One of the points about forecasting Montier makes is that "The bottom line from this whistle-stop tour of the failure of forecasting is that it would be sheer madness to base an investment process around our seriously flawed ability to divine the future."- However, there appears more utility to forecasting, then would be merely the actual prediction element, The following highlights and summarises some of the responses from people to this discussion on my 'Linkedin' group.

Firstly some very inciteful responses which I am paraphrasing from Can Esenbel:
- Forecasting can make you prone to start caring more about being right than about making money.
- Forecasts can become 'Hopecasts' or 'Wishcasts': 'The other problem with a forecast is that we will tend to integrate our inside view as an outside view'.
- Forecasts can cause us to lose sight of the context: 'We will tend to overweight recent, eloquent confirming information while ignoring or downplaying info that we don't like. We are also like to compress or expand time frames to manage our views'.

Durga G added to this that a danger with forecasts is that:
- It is how we respond to forecasts that often be the problem : Forecasts are fine but it is the certainty one attaches to a forecast that leads to trouble.

Biju Dominic made an interesting point:
- It is inherently difficult for a human beings to believe that the future is uncertain. So despite enough and more studies that show that it is impossible to predict the movements of a financial market, or the way the consumer will behave in the market place (I am sure many of us have read the book Why Most Things Fail by Paul Ormerod and Wrong, Why Experts Keep Failing us by David H. Freedman) it is imperative that 'experts' are expected to have a point of view about the future.

- Ian Copsey, a market forecaster himself, backed-up this point of Biju's with some examples of his work. Ian is the sort of person Biju was highlighting, experts in their field, who may be able to add some value to the traders/ investors own perspective.

- Some excellent counter-points were however made by Richard Brown and Mao Song-Gong. Which brought in some other aspects of forecasting: The essence of these being that despite most (not all) forecasts being nothing more than educated guesses with little chance of being accurate, they do provide a context or focal point for a trader or investor, without a forecast or belief many people are really just looking at numbers moving. Further to this, they also provide an opportunity to learn and increase our knowledge of the current market, by having a reference point to assess what we see against what we think. Thus they help us in proving clues which can help us to try and complete parts of the jigsaw: Taking this analogy a little further, when one tries to complete a jigsaw, it is often a trial and error process, which involves looking for pieces that may fit. - In the markets of course it is virtually impossible to ever complete the jigsaw, but at least one may try and put parts of it together to have a sense of the bigger picture.

I would like to add something to this, which partly builds upon the point made by Biju above. Markets are inherently uncertain places, and entering them means venturing into an uncertain world. - We need a strategy to cope with this, or we risk being paralysed by fear, and thus never taking part in seeking risk. - These strategies will depend very much on ones personality and behavioural characteristics and preferences. -Thus an inherent part of your strategic approach to trading will be an attempt to reduce your uncertainty (or ambiguity aversion). The tactics you adopt to achieve this will probably depend on your preferred behavioural approach or personality: For some it will be following rules, or taking a systematic approach; others will like to expend energy keeping very close control on themselves, their trading and their risk-management, others will limit their exposure to the market to mere seconds or minutes, - I could go on, but I am sure you get my point. - One approach to reducing uncertainty is to make (or to have) a prediction of what may happen next, whether or not it turns out that way, is not really as important in this sense, as to how it reduces uncertainty within the trader's mind.

Further to this last point, and to echo some of the above points, forecasts also provide a reference and focal point to events in the market. Some traders are able to make successful trades out of incorrect forecasts, this is because the subsequent market actions following the forecasts enables the trader to eliminate certain courses of action, and can help highlight favourable risk/reward set-ups.

In summing up, I would say that it is not folly to forecast, there is real utility to forecasting and to reading peoples forecasts, though this will largely depend on a trader/investor's working approach and style. I would however add a note of caution; traders and investors must make sure they keep forecasts in perspective, they are not and never will be a road-map of the market, in most cases they are 'best-guesses' based off limited and ever-changing variables. However they are useful noise, knowledge and information, they can help provide a focal point to the market and a context to price action and news/events. Further more they will always be needed and demanded because people need ways to reduce uncertainty, and will seek leadership and expert advice. .

The folly of forecssting

“It’s tough to make predictions, especially about the future” -  Niels Bohr.

"Those who have knowledge, don't predict. Those who predict, don't have knowledge", Lao Tzu, 6th Century BC Chinese Poet

As someone who enjoyed trying to make market predictions (‘trying’ being the operative word), I just wonder how much time, effort and energy is wasted in what Nassim Taleb refers to in ‘The Black Swan’ as the "the scandal of prediction". Taleb goes onto to talk about those engaged in prediction as suffering "epistemic arrogance" and "self delusion", and in classic Teleb style, he calls these experts “people who do not know what they do not know".

In his excellent ‘Little book of Behavioural Investing’ James Montier talks about those analysts who try to predict the target price of stocks, noting that not only was the average price of their predictions often way off the mark, but in the years from 2000 to 2008 the analysts only managed to actually predict direction right on four occasions out of a possible nine. 

I omce read an article from Louis Gave of Gavekal research, ‘According to a 2002 IMF study, out of 74 identified episodes of recession in different countries, only four had been correctly predicted by econometric forecasts published just three months prior the recession year - and in two-thirds of cases, consensus economists had failed to 'forecast' the recession even four months after it had started."

Finding Value v Predicting Markets.
In my own trading career. (1986 to 2009), I spent many years trying to predict markets. - I became admitdely very good at finding value. That is not the same as prediction. Finding value, is identifying opportunities whihc may be underpriced or undervalued, or which have a favourbale risk-reward structure which fits in with your overall trading strategy. many people confuse 'finding value', with 'predicitng market direction'.   

A trader's job is to find value and then monetise that value. - 2 different jobs in themsleves. A traders job i not to predict where the market is going. Analysts do that, som ebetter than others. 

When I traded often I may feel the market is going higher. - I might share this with colleagues or a manager. Once my manager came and commended me for a good call. he then asked how much i made from it. - feeling slightly embarresed, I said nothing. He asked why. I said the risk-reward set-up at no point suited me.. There was no value for me in ging long, I didnt have a stop level I was comfotable with, and so I missed the move. 

However, on another occasions, the (almost) the opposite happened. I said I felt th emarket was heading higher, but without a strong convition. That week the market dropped sharply. - He I replid nothing, i woudl predict the maket is i sometimes,