Louis, You are welcome, I am very happy to be of assistence.
As far as backtesting sentiment data, it is pretty straightforward. Download the data off the internet, and import it into Ami as a textfile. You can then use it in backtests as you would any other instrument. Do a google search on 'sentiment data' and you will find a wealth of information. If you are interested in trading forex, I forgot to mention that OANDA offers near real-time snapshots of aggregate client positions. I experimented with this data at one point and found to my suprise (well not really) that the small OANDA traders are pretty bad as a group. You can use this data as another source of research ideas if that interests you (I shelved the project myself to focus on equities). I remember I tested something like either buying or fading breakouts based on which way the small-money was leaning, and found pretty good if preliminary results from that. It is definitely an interesting avenue for you or someone else out there, I believe. I can share with you a few of my models, so if you want send me an email we can talk more. However, I highly recommend you spend a few days (or more) googling on your own for the resources I discussed in my last post. You should really want to take ownership of this process as much as you can. It won't be easy at first, but once you build a foundation, hopefully a certain type of trader's intuition will start to kick in for you. You'll know it when you feel it, is how I can best describe it. --- In [email protected], "Louis Préfontaine" <[EMAIL PROTECTED]> wrote: > > Thank you tinga_tinga for your answer. This is like gold to me. This is > really what I have been hoping for. Just one little question: how do you do > to backtest data like market sentiment, bull-bear type investor > surveys? > > Thanks a lot for the tips; I will work on that! > > Thank you to you Howard too. I am reading your book and it helps me to get > the maximum out of Amibroker! > > Louis > > 2008/3/25, tinga_tinga <[EMAIL PROTECTED]>: > > > > Louis, > > > > I have been an Amibroker user for over 2 years now, and wanted to > > offer a few ideas for you to think about. I trade equities almost > > exlusively, so my indicators are geared towards stocks more than > > commodities. Here is one trader's brain dump :) > > > > 1. Moving averages work well for me. If you backtest pullbacks > > (defined however you wish, by % change or by RSI level or whatever > > other indicator you wish to use, not important) you will see how > > stocks act differently whether they are trading above or below > > certain moving averages, say a 200 day SMA, for instance. I think > > this in itself will give you something to build upon, and open up a > > world of trading ideas. Larry Connors has some good research in this > > area, while I find his offerings a little too promotional for my > > liking, you can use his ideas and backtest them yourself as I have. > > > > 2. Don't resrict yourself to an individual instrument or index, think > > about intermarket relationships. To that end, try to get your hands > > on ANY Ned Davis Research charts you can find in the public domain. > > It will open another world of ideas in regards to things such as > > moving averages of breadth indicators, to give just one example. > > Also grab "The Quantification Predicament" which earned NDR analyst > > Timothy Hayes the 1996 Charles H. Dow Award. > > > > 3. Backtest and use moving averages in relation to sentiment > > indicators such as put/call ratios and the bull-bear type investor > > surveys. > > > > 4. If you want to trade breakouts, think and backtest whether it > > makes more sense to buy at the point of a breakout, or after a > > pullback to the point of a breakout (or thereabouts). I have found > > my success with the later method. > > > > 5. You can combine both a trend following and a mean reversion system > > using moving averages and/or breakouts and achieve great results. It > > is just a matter of timeframes. Alexander Elder talks about this in > > his books. This trend/reversion fusion is the core of my individual > > stock selection method. > > > > 6. Get on the Social Science Research Network and find what you can > > in regards to earnings estimate revisions, momentum, and industry > > momentum (aka - does the group explain the stock momentum anomaly???). > > > > 7. Be curious, persistent, enthusiastic, and above all never stop > > imagining new research paths (highs/lows, advancers/decliners, % > > changes, sentiment, analyst revisions, insider buying, COT reports) > > it's all out there waiting for you to backtest :) > > > > Good luck! > > > > --- In [email protected] <amibroker% 40yahoogroups.com>, "Louis > > Préfontaine" > > <rockprog80@> wrote: > > > > > > Thank you Brian for your reply and thank you all for your replies. > > I was > > > not looking for an easy answer, but rather what people thought about > > > Breakouts and moving averages as main indicators. Of course I will > > build my > > > own system, but I believe knowing what people with more experience > > are doing > > > can help. > > > > > > Louis > > > > > > 2008/3/24, brian_z111 <brian_z111@>: > > > > > > > > > > Louis, > > > > > > > > As long as you understand I don't deal (intentionally) in the base > > > > human emotions (they have no value to me). > > > > > > > > My philosphical points were addressed 'To Whom It May Concern'. > > > > > > > > People are totally mistaken if they think there is no relevance to > > > > trading in my 'philosphical' post, however few are interested and > > I > > > > don't have the time to do the subject justice anyway. > > > > > > > > My 'easy' answer to you was c) none of the above, but, IMO the > > > > problem is that you don't have the experience to make the > > necessary > > > > distinctions for yourself (is it my fault that you are in such a > > > > hurry?). Also, it is not a good thing to 'lift' your trades from > > > > others. People trade best wnen they develop their own > > systems/styles > > > > etc so that is why I gave you that answer - it is in your best > > > > interests. As well as that, good opinion on this subject abounds > > in > > > > the forum and the books you have purchased e.g. Howard has given > > his > > > > opinion on the importance of people choosing their own Objective > > > > Function and the fact that his recommended approach sidelines > > trading > > > > pyschology, so it is not as if the question went unanswered. (I > > > > understand that my esoteric answers are not to everyones taste, > > > > perhaps not yours, but there are plenty of other people prepared > > to > > > > give an opinion. > > > > > > > > However that is just my answer - there are another 7000+ people > > who > > > > can give you any answer they like. > > > > > > > > Cheers, > > > > > > > > brian_z > > > > > > > > --- In [email protected] <amibroker% 40yahoogroups.com><amibroker% > > 40yahoogroups.com>, "Louis > > > > Préfontaine" > > > > <rockprog80@> wrote: > > > > > > > > > > Ok I was naive; I thought I could simply ask and get some easy > > > > answers which > > > > > would get me right on track to get new ideas on how to build a > > > > hourly > > > > > system... > > > > > > > > > > I'll stick around and try to get some piece of wisdom and maybe > > I > > > > will be > > > > > able to build something with those that will eventually allow > > me to > > > > build > > > > > the system I want to build! > > > > > > > > > > Louis > > > > > > > > > > > > > > > 2008/3/21, brian_z111 <brian_z111@>: > > > > > > > > > > > > > > > > (Subjective) investigations into the 'human condition' have > > been > > > > > > going on, in parallel with our search for objective truths, as > > > > long > > > > > > as humanity has been around. > > > > > > > > > > > > This body of information has been collected and preserved, by > > the > > > > > > few, for the benefit of mankind (the many) and consitutes a > > > > SCIENCE > > > > > > to its guardians, adherents and students. > > > > > > > > > > > > From that body of WISDOM two principles can be extracted that > > are > > > > > > relevant to your comments: > > > > > > > > > > > > holism is universally persistent (all things are made in the > > > > IMAGE of > > > > > > the creator) > > > > > > > > > > > > and, > > > > > > > > > > > > flowing from that, we derive the principle of CORRESPONDENCE > > > > > > (operating principles in one sphere, have their corresponding > > > > > > principle in another)... > > > > > > > > > > > > ...but that is going to far OT. > > > > > > > > > > > > Over to trading (OR how the above relates to trading): > > > > > > > > > > > > Over the long term the bias of the (stock) market is a > > function of > > > > > > the earnings performance of the component companies. > > > > > > > > > > > > This is skewed by the behaviour of market participants, which > > > > > > introduces randomness to the markets. > > > > > > > > > > > > The shorter the timeframe the more dominant is 'randomness' > > > > (which of > > > > > > course is not true randomness). > > > > > > > > > > > > (If you are interested in the subject of organising > > principles and > > > > > > holism Carl Jung's work on Psychological Archetypes is a > > wonderful > > > > > > example of how the universal paradigms play out in the > > affairs of > > > > > > wo/mankind). > > > > > > > > > > > > brian_z *:-) > > > > > > > > > > > > > > > > > > --- In [email protected] <amibroker% 40yahoogroups.com><amibroker% > > 40yahoogroups.com><amibroker% > > > > 40yahoogroups.com>, "Ronald > > > > > > Davis" <xokie7@> wrote: > > > > > > > > > > > > > > I maintain the view that algorithms exist in nature, and > > that > > > > > > people who develop algorighms are only discovering another > > one of > > > > > > nature's secrets. > > > > > > > > > > > > > > When my son first showed me Amibroker several years ago, I > > > > looked > > > > > > at charts with Stochastics, and RSI, and I became convinced > > that > > > > > > mother nature has algorithms that can find the central core of > > > > all of > > > > > > that volatility. > > > > > > > > > > > > > > I have yet to discover mother nature's algorighms, but my > > > > attempts > > > > > > have led me to some conclusions. > > > > > > > > > > > > > > -------------------------------------------------------- -- > > > > > > ---------------------------------------------------------- > > > > > > ------ > > > > > > > > > > > > > > My results WERE BEST when I "AVERAGED THE LAST SEVERAL > > HUNDRED > > > > DAYS > > > > > > OF ACTIVITY" > > > > > > > > > > > > > > and watched the LAST 9 DAYS>of the performance of this > > average > > > > of > > > > > > hundreds of days. > > > > > > > > > > > > > > -------------------------------------------------------- -- > > > > > > ---------------------------------------------------------- > > > > > > ------ > > > > > > > > > > > > > > My results WERE LESS GOOD when I "AVERAGED OF LAST 9 DAYS OF > > > > > > ACTIVITY" > > > > > > > > > > > > > > and watched the LAST 9 DAYS>of the performance of this > > average > > > > of > > > > > > only 9 days. > > > > > > > > > > > > > > -------------------------------------------------------- -- > > > > > > ---------------------------------------------------------- > > > > > > ----- > > > > > > > > > > > > > > Hope this helps someone. Ron D > > > > > > > > > > > > > > -------------------------------------------------------- -- > > > > > > ---------------------------- > > > > > > > > > > > > > > ----- Original Message ----- > > > > > > > From: Louis Préfontaine > > > > > > > To: [email protected] <amibroker% 40yahoogroups.com><amibroker% > > 40yahoogroups.com><amibroker%40yahoogroups.com> > > > > > > > Sent: Thursday, March 20, 2008 10:25 PM > > > > > > > Subject: [amibroker] Philosophical question > > > > > > > > > > > > > > > > > > > > > Hi group, > > > > > > > > > > > > > > I just began reading Howard Bandy's book (even though I did > > not > > > > > > finish Aronson's book yet...), and a somehow philosophical > > > > question > > > > > > came to my mind when he speaks about the market's > > inefficiency and > > > > > > how we must take advantage of it. He talks both about moving > > > > > > averages and breakout, and I was wondering which one of the > > two > > > > > > techniques do you think is the more promising for such a > > system? > > > > > > > > > > > > > > I ask this because as far as subjective technical analysis > > is > > > > > > concerned, I am more used with breakout techniques. But the > > real > > > > > > inefficiency in breakout techniques comes from time, that is > > if > > > > one > > > > > > can act quickly enough to make a profit from the sudden > > change in > > > > > > price. But from my experience it seems to be more difficult > > with > > > > EOD > > > > > > or hourly data. And it is less profitable for someone (like > > me) > > > > who > > > > > > is using options, which tend to anticipate the change quicker > > > > than it > > > > > > really happens. > > > > > > > > > > > > > > Moving averages techniques, on the other side, seems a bit > > > > > > mystical to me, and maybe a bit too simple or too « easy ». I > > > > don't > > > > > > know much about them... > > > > > > > > > > > > > > But anyway, my question is: which one of those two > > techniques do > > > > > > you prefer, or do you use both for entering a trade, or > > shorting a > > > > > > trade? What can be a good way to trade for someone (like me) > > who > > > > > > wants to trade hourly data and can't always get the beginning > > of a > > > > > > breakout? > > > > > > > > > > > > > > Thanks! > > > > > > > > > > > > > > Louis > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > >
