Dow

[evilroyrule]

asia pumping up the jam tonight. nikkei 12 month high.

[winner69]

Yipee ...... DOW back to 11000

On and upwards from here to 12,000 and maybe 13,000

But gee the DOW was 11000 in 1999

Todays 11,000 just what you expect in a secular bear market .... and we'll probably celebrate again in 2014 and 2015 when the DOW reaches 11,000

[beacon]

Todays 11,000 just what you expect in a secular bear market .... and we'll probably celebrate again in 2014 and 2015 when the DOW reaches 11,000

Et tu Brute'

[Phaedrus]

[Phaedrus]

It seems that more people on ST are now applying Fibonacci levels to indices and stocks in an attempt to identify future turning points. In view of this, I thought it might be interesting to "backtest" Fibonacci levels over 20 years of Dow history and see how well they identified the significant turning points of this period.

Trends are followed by countertrends. Fibonacci theorists claim that the extent of these retracements is often a Fibonacci ratio of the extent of the previous move. The countertrend "ought" to retrace 23.6% or 38.2% or 50% or 61.8% or 100% or 123.6% or 150% or 161.8% (etc!) of the previous major trend. You can see from the charts below that after each and every major trend, Fibonacci levels failed to identify (let alone predict) the extent of the following retracement, even approximately.

Troubled by too many significant highs and lows failing to occur at conventional Fibonacci levels? No problem! As illustrated in the latest chart (bottom right) all you need do is simply apply another set of Fibonacci levels within each of the original ones. Such a subdivision gives the current uptrend 26 putative reversal levels - so far. This technique (as posted right here on ST) gives such a multiplicity of possible reversal levels that it virtually ensures any event deemed to be of significance will be "at or near" a Fibonacci level.



A recent study by Professor Roy Batchelor and Richard Ramyar found no evidence that Fibonacci numbers work in American stockmarkets. They tested nearly 90 years of history of the Dow Jones Industrial Average (1914-2002) and found no indication that trends tend to reverse at the 61.8% level - or at any other Fibonacci level.

http://www.cass.city.ac.uk/media/sto...004_65846.html

[peat]

maybe you should have started a new thread.
I think perhaps you're taking the fib claims a little too literally in that no one claims its perfect but that there are a lot of instances where they act as support resistance levels which may or may not hold.
There are so many examples of them being useful that your 1 post denigration using one index isnt really a satisfactory refutation in my opinion
Also you didnt mention the fib that I prefer most of all in conjunction with the .618 fib, and thats .786 (the square root of .618)

There - I did bite! (I was very tempted not to)

[STRAT]

What? only one nibble.
Come on you Fib fans. Get stuck in and defend the cause.

Peat reckons they may or may not work which is a bit like it may rain tomorrow or it may not. Not exactly a strong defence

[peat]

well it seems to me when I draw the Dow Fibs that is there IS some magical attraction to fib levels during the correction - the price corrects to the 23 the 50 and 61.82 levels.

[peat]

and on the way down the 2.61 extension is just a little uncanny dont you think?

[Phaedrus]

Peat, our plots differ because we are basing our respective Fibonacci retracements on different starting points. The attached chart shows just a few of many possible start date options. All are equally valid, and of course even more become available as you extend the timespan covered by the chart. The low that you elected to use is marked by a red arrow, mine light green. Fibonacci levels are colour coded to match their respective start point arrows.

This chart gives some idea of the very large number of Fibonacci retracement levels that can be generated. They are so closely packed that one can be found for nearly every peak and trough. I have included Peat's favourite 78.6% fib. You can see that it is no more effective or accurate than any of the other retracement levels. All these lines - and not one got it right! I certainly see no "magical attraction to fib levels" and the only thing I find "uncanny" is that the reversal, when it came, was between Fib lines - and in an area where they were unusually sparse!