Hi ACE,
Our posts crossed. Our AB starting profiles are very similar, in more ways than one. Extending my previous post a little: Some say that AB has the best BackTester going (especially with its portfolio capabilties). I wouldn't focus on that point. If I had to sum up AB I would say that it is the most customizable software around and for trading survivors that is the most important criteria. One of the hardest things for me to do, and I think it applies to everyone, is to divest myself of my fixed beliefs e.g. when I started using AB my user paradigm was that the solutions I required were generic and that they were in AB somewhere ... they must be in there because Tomasz is no fool, and he has beeb around for years, so he must know what we all need and therefore he has incorporated it, or will do so, as soon as time permits (wrong!). As an aside the incorrect belief that I have found hardest to shake was the view that there is causality in the markets, say, stockmarkets, in the short term time frame ... the corollary to this is that time and time again I attempted to model market structures, which IMO is an effort doomed to failure (please don't ask me to explain that statement ... from time to time I post a few hints aobut it, in various places, and I really can't do more than that). > " start with RMath from the get go" > . What do you mean by this statement? Statistics for traders is one of my specialities ... I just don't go about it in a classical or academic way. I didn't know that when I got started. Once this fact started to dawn on me, somewhat dimly at first, I started to take notice of the people in the forum who were specialists and knowledgeable in the field. Howard and Patrick are the main two. Patrick doesn't post often but he was a kind of silent mentor in a way in that he is a good community member who has sponsored several AB stats initiatives (plugins etc), so I was able to progess of the back of his efforts without any exenditure of energy on my part. Recently he donated a RMath plugin to the community (see the file section) but it is more than just the plugin ... it represents the progress that Patrick made in his years of trading/thinking about analysis and stats applications. So it is an off the shelf solution that fits me perfectly ... Patrick said that anything we can't do in AB we can do very well in R (plus there will be more to come as more of us start to use it and share info via this board). I would also be better off if I had learnt a second language so in my case I would have been better served if I had discovered and learnt RMath 2-3 years ago (as I said ... you need your specs to find your specs). In the past Patrick talked about Matlab, and I had a quick look at it, but at the time it seemed like a bridge too far (Patrick was a stats God in the sky) ... as I said, "an apple doesn't ripen overnight". RMath suits me because I see the foundation of trading as analysis, and the big advantage of R is that it is purely an analytical tool (one with a strong academic following too). I am not so interested in customized implementation at this stage i.e. pushing a button and making AB behave like my own trading platform (which it can do if that is what you want). > "read Ralph Vince earlier" > > Thank you for the suggestion of this author. Which Books of his >would you > recommend for me to start with first. He wrote four books ... IMO the first is mandatory reading. Vince - Ralph, "Portfolio Management Formulas", John Wiley & Sons, Inc, 1990. http://zboard.wordpress.com/library/books/ Or see Howards resource page http://www.quantitativetradingsystems.com/resources.html Don't throw it away if it seems too hard at first ... keep it handy for a few years. (Someone told me to "go read Ralph" years ago and I didn't follow through .. still there are only so many hours in the day). --- In amibroker@yahoogroups.com, "caternore" <catern...@...> wrote: > > Blair > Thank you for your suggestion. I will definitely put that down on my list > of important things, especially 1 and 3. > > sidhartha70 > Thanks for your Suggestion. Understanding AB execution cycle will be added > to my list. Again thanks > > Herman > "You'll learn to love afl for quick research and system explorations" > > Thank you for your words of encouragement. I have never had to deal with a > Trading Program and Language as complex ( or robust if you choose) like AB > and AFL and I do not have any background in programming. I look forward to > the day where that statement rings true for me as well. > > brian_z111 > "That's not a good question ... it doesn't appear to get to the heart of your > enquiry ... perhaps you had difficulty articulating your 'need' or maybe there > isn't a real question in there?" > > Brian you are correct, I do have difficulty articulating my needs. You > see the problem with me is that I have no background in programming, I have > never worked at a trading firm or on the floor of any exchange. Add to > that this is my first time dealing with a complex program (at least to me) > like AB. This being said it may seem difficult for me to articulate exactly > what I want or what I am looking for, but This is how I learn, I start from > the most generalist view point that I can understand and then break that > down to more complex Ideas(structures, Etc) . I also have found that when > you ask a generalist question you may be pleasantly surprise the direction a > conversation may take and you may end up learning more about the topic at > hand then you would have by asking a more specific question. Over all it > seems as if you general got the gist of my question and answered > appropriately. > > "For me learning AB is probably more difficult than developing trading > philosophies etc ... my 'good ideas' in tray is overflowing and a long way > ahead > of my ability to implement them (assuming I have the time)." > > I agree with your statements. (at least for now). I have so.... so many > ideas, that I would love to validate through Amibroker. I must admit before > I purchased AB, my mind was hooked on to (what seems to be) a more easier > program to operate and learn( Traderstudio, Trading bloxs, etc). While > these programs seems to have some exceptional qualities to them. When > comparing price points vs. ability and the fact that one would still have to > learn a program language to implement unique ideas. I think that (at least > for now) Amibroker is a good deal and worth learning how to use effectively. > I have taken your suggestion to heart. instead of trying to learn > everything about AB, I will take the Piecemeal approach while still trying > to learn AB in and out. > > "Perhaps learn RMath and try to use that for the above." > > Rmath seems to be very Interesting. I look forward to researching this. I > have a question for you regarding your statement > > " start with RMath from the get go" > > . What do you mean by this statement? > > "read Ralph Vince earlier" > > Thank you for the suggestion of this author. Which Books of his would you > recommend for me to start with first. > > Ok Guys I definitely appreciate you guys answering my questions. From this > post (and other I have posted) and from reading various other post on this > board. I have come up with my to do list. > > 1) Read AB User guide, KB, UKB, Yahoo board, etc. > 2) Learn how AB works with arrays > 3) Understanding Variables, then expanding to static, dynamic, and persistent. > 4) Understanding AB execution cycle. > 5) Researching Batman and seeing how it would affect the way I use AB > 6) Read some Ralph Vince > 7) Researching and learning Rmath > > This list is open to anyone who would like to expand this list or add a view > point. > > Again Thanks > ACE > > > > > > > > > --- In amibroker@yahoogroups.com, "brian_z111" <brian_z111@> wrote: > > > > >If you guys knew what you Knew now when you first started using > > >>Amibroker, what would you do differently? > > > > That's not a good question ... it doesn't appear to get to the heart of > > your enquiry ... perhaps you had difficulty articulating your 'need' or > > maybe there isn't a real question in there? > > > > Possibly you are asking: > > > > 1). - What mistakes did I make, that, with the benefit of hindsight, I > > would avoid now? > > 2). - OR, What would I use AB for if I was starting again, with the > > knowledge and experience that I now have? > > > > > > > > > > > > 1). - What mistakes did I make, that, with the benefit of hindsight, I > > would avoid now? > > > > > > There is no answer to this question because my mistakes made me who I am > > today ... as a trader, AB user and AB community member. > > > > You have to know where I started to understand that the mistakes were my > > own, and necessary. Also, hindsight is 20/20 vision .... you need your > > specs to find your specs etc. > > > > I think there are two components involved; learning AB's capabilities and > > developing your own trading style, and strategies, and the latter > > determines which of AB's capabilties you are going to specialize in. > > > > The first is an ongoing issue for me ... there is so much involved and > > there is no 'mother of all books/indexes' ... I am the kind of person who > > just wants to reach for the index, find the reference, learn the relevant > > portion and then continue on my way (all in the privacy of my own office). > > > > The mother lode just aint there, not in one place and not in its entirety > > ... probably AB is too big now to put it all in one place anyway. > > > > A lot has changed since I started using AB, around 3-4 years ago .... > > Tomasz has done a lot, publically and privately, to provide more learning > > assistance for newcomers. > > > > For me learning AB is probably more difficult than developing trading > > philosophies etc ... my 'good ideas' in tray is overflowing and a long way > > ahead of my ability to implement them (assuming I have the time). > > > > Which came first ... the chicken or the egg? > > > > - learn AB in a broad brushstroke way and then start work on making a > > living and specialising in one aspect of AB? > > > > - OR just learn what you need to know, to implement your ideas, as you go > > along (a piecemeal approach). > > > > I have used the second approach. > > > > IMO a lot of people end up becoming AB professionals, instead of trading > > professionals, so I am wary about going down the first path, but I am > > biased. > > > > Another common mistake is for developing traders to become algorithm > > addicts or IT addicts .. once you are hooked on them you can forget about > > trading and become a consultant instead. > > > > So, looking back the main issues seems to be how I went about learning AB > > compared to the most efficient way of doing that ... I could have done > > better in that regard but it wasn't all my fault. > > > > > > > > > > > > > > 2). - OR, knowing what I know now what would I use AB for if I was starting > > again, with the knowledge and experience that I now have? > > > > > > I really only use AB for: > > > > > > Data: > > - access free Yahoo EOD data, via AmiQuote, for researching my ideas > > (absolute accuracy of the data isn't crucial). > > - access RT data from eSignal (because it is reasonable and it has > > international exchanges available ... I live outside the US). I mainly use > > RT data for R&D rather than live trading. > > > > > > R&D: > > > > - I only use basic customisation of charts > > - I use indicators and explorations for system design and testing (the > > ideas are subtle and, sort of, complex but the implementation is basic) > > - I sometimes use scan mode for R&D or finding a list of stocks for > > potential trades > > > > So, starting today, with a clean slate and being brutally minimilistic: > > > > - don't answer posts > > - don't get involved in community effort > > - don't publish anything, anywhere > > - forget about all the other AB features except for charting, scanning and > > explorations > > - one way or another develop my own backtester, and other stuff, inside or > > outside of AB (for personal use only ... if it was good enough, and of > > interest to other traders I would give it away somewhere except for my core > > trading secrets) > > - Perhaps learn RMath and try to use that for the above > > - maybe I won't do any more extensive R&D ... just trade to pay the bills > > and go to the beach more often. > > > > Possibly the only advanced feature in AB that might tempt me to further > > learning is AutoTrading or, as an outside chance, Optimization. > > > > If I had my time again possibly I would: > > > > - read the manual. > > - read RalphVince earlier > > - start with RMath from the get go > > - try not to get angry about the things that pinged me off > > > > I believe that I have squeezed the juice out of the TA orange, so I don't > > need indicators, optimization etc. > > > > The only thing that might change, in the future, is if I decide to try > > another style, which would be what I call 'clip trading' e.g. spread > > trading, mispriced options etc .... that is to say 'arbitrage trading', > > which I believe is a viable style but one that I have avoided because I > > would have to compete with the big boys on their turf (high speed, high IT > > skills required). > > > > I guess I can't blame AB for all of that! > > > > > > > > > > --- In amibroker@yahoogroups.com, "caternore" <caternore@> wrote: > > > > > > Hello All, > > > I am new to Amibroker. Amibroker seems highly customizable and robust. > > > The community on this board seems to represent various different Ideas, > > > view points and user knowledge(trading, Amibroker, etc) . My question to > > > the board If you guys knew what you Knew now when you first started > > > using Amibroker, and or first started trading with Amibroker, what would > > > you do differently? > > > > > > Thank you > > > ACE > > > > > >
