Do you see patterns in Random Walks?

Discussion in 'Technical Analysis' started by atlTrader666, Aug 10, 2011.

  1. I'm interested in the more esoteric technical analysis and why people believe it (against all evidence)... Aside support and resistance and some momentum patterns, the rest of TA seems like nonsense. Academics have tested TA strategies for half a century and they all call it bullsh1t. There is vague proof of short-term momentum. Only Benoit Mandelbrot discovered something valid: volatility tends to cluster. You can't make money of that though since traders have intuitively known and incorporate higher premiums in their options trading when volatility increases.

    My question: If you were to create a random walk in Excel do you think you would differentiate the chart from say a stock market chart?

    Seriously guys Google image a random walk chart or create one in Excel and you will be amazed at the amount of double tops/bottoms, bull/bear flags, breakouts from consolidation areas, etc. that you will discover. Even Fibonacci lines will look like viable entry & exit points.
     
  2. Roark

    Roark

    A picture is worth a 1,000 words. Why don't you post a couple?
     
  3. wrbtrader

    wrbtrader

    I've seen many "random walk" charts. They look similar to any other price chart via swing points and reaction points only. Yet, they have weird or strange looking volatility expansion, weird or strange looking volatility contraction along with poor distribution of price ticks when the lows and highs are different from the close or open in an interval (see my P.S. statement below).

    Simply, when I see those "random walk" charts amongst real charts...the random walk charts are obvious and the first thought I have is fake charts.

    Yeah...I have a few buddies heavy into the random theory and they seem confused every time I'm able to identify accurately which charts were generated via random prices and which charts were the charts of real stocks, futures or forex.

    My point is that there's more to the price action than just double tops/bottoms, swing points, reaction points, flags, breakouts or whatever. Simply, these are not trade signals all by themselves. They are just price areas of interest that should prompt a trader to take a closer look. In contrast, the trade signals is what's occurring within those patterns you and I have mentioned and that's something my buddies can't understand...including most academics.

    That's why they had poor trading results because they think a double bottom is a Long signal all by itself. It's not. Thus, there's more to TA than that and the few that realize such will be the ones that benefit from using TA.

    P.S. I can see the difference between random price charts versus real charts when the data is presented as candlestick charts or bar charts. Yet, I can not see the difference when the data is presented as line charts, dotted charts or charts where the intervals have been so compressed that you can't determine if it's candlestick charts or bar charts...charts typically not used by those that are traders.

    Mark
     
  4. Yes, I see patterns on Random Walks. When I go out for a walk at night --- doesn't matter where I'm going --- the sky appears dark. When I go walking during the day, the sky appears bright.
     
  5. etile

    etile

    I don't understand the random walk thesis. If you expand your time horizons far enough (to years) you can see definitive trends. From a long term perspective prices will follow fundamentals of a stock. Sure the premium multiple will probably change over time, but a good stock or bad stock will show its innards eventually and this will be reflective of its price that is often arrived at in a trend.

    I'm not a random walk expert, or even that smart, but it seems that the random walk theory presuppose that all market participants act randomly. Yet if you think about, every trader is trying to think about what the masses as a whole will do. The confluence of the same sentiment occurring over a prolonged period of time does not happen by happenstance.

    Maybe its just me.
     
    Wisard likes this.
  6. If you flip a thousand coins, you will get long runs of consecutive heads or tails. These would appear to be trends, but they are in fact random and they do not change the odds of the next throw. With is 50/50.

    The market is the same.
     
  7. the1

    the1

    My entire Master's Degree -- Statistical Finance -- is based on the premise that the markets is one gigantic random number generator and 95% of the tools I use are statistical in nature but at the same time, I've stood toe to toe with traders who have come to the industry with backgrounds from the likes of Computer Science or Sociology and wouldn't recognize an Autocorrelation Process if they tripped over it but managed to trade the markets quite competently, nonetheless. To answer the question....does a true random number generator produce patterns? The answer is most definitely yes. I've designed many of them and they have all produced trends, patterns, and chaos. They even produce levels of support and resistance that you would swear are purely psychological if you didn't know better.

    So are the markets random? No one will ever be able to answer the question without doubt but it's really not that important because trading profits are more strongly correlated to the talents of the traders, not the design of the system.
     
  8. etile

    etile

    This presupposes that each trader makes a each and every trade in a vacuum, which isn't true. Moreover, each and every trade on the tape would be completely random. The random market theory seems to say that if you put a bunch of people that a single idea is incapable of capturing the minds of the majority. The rise in the price of gold seems to say to the contrary.
     
  9. the1

    the1

    You can develop a very simple simulation like this using excel. Just use the rand function with a tick size of 1 with three possible outcomes: -1, 0, +1. Start with a number such as 1000 and add a sequence of something like 10,000 outputs and pull up a chart. Throw a 9 period and 26 period overlay MA on the output and pull up a chart. Bingo! What you will be looking at will be a stock chart with a MACD on it. It's pretty amazing how such a simple model such as this will look exactly like a stock chart.

     
  10. #10     Aug 10, 2011