Welcome to the new Traders Laboratory! Please bear with us as we finish the migration over the next few days. If you find any issues, want to leave feedback, get in touch with us, or offer suggestions please post to the Support forum here.
andypap
Members-
Content Count
52 -
Joined
-
Last visited
Content Type
Profiles
Forums
Calendar
Articles
Everything posted by andypap
-
i am really surprised that almost everyone thinks that this is somehow a fair game. consider these three factors inside information, manipulation and extremely friendly legislation. please let me know who on earth who has enough power will not use it to have an advantage
-
if any of you is intrested check the secret life of chaos bbc documentary in youtube. it is all about how selforganization , feedback and evolution can create complex systems whithout any initial purpose in doing so even intelligent life the analogy to the markets and the timeframes down to the tick is more than obvious. if some of you are intrested we can start a meaningfull conversation about strategies and performance stats that can answer the question that started this thread and many other questions by deduction.. there are quite a few good programmers and traders here and if we openly open our "equity charts"(not our code or setups just our performance) to each other we consitute quite a good sample of what is likely possible. apart than that whitch in good faith will be accurate data all other conversation of this sort or this trading room and that idea and mp and fibonacci vortex and outter space is politely speaking speculation and wishfull thinking or plain horshit because they are not backed up by any kind of EVIDENCE.. is engaging in verbal arguments and various hypothesis or mere fantasies the reason for you to be here.. i havn't seen one single thread of evidence for what is claimed in various threads as a sound methodology so far. post your acount progress if you are descreationary traders or your performance stats if you design strategies and lets start a real conversation about the markets...
-
its all about position,capitalization and personal risk of ruin to determine the players and their focus and perception of future value to determine the directional bias these are the variables at play everybody claims how the markets have change recently well nothing has changend i think that this is a claim as ludicrous as saying the world has changed and there will be no more war and the rich will donate their money to the poor... if you play by the rules as a small investor you will get your banana if you develope your skills to the max as a day trader you will get your banana
-
hi any suggestions on where to start for computing the correlations between the equity curve of a strategy and predefined popularions of variables,,, or even a group of equity curves of the same strategy among different data sets and sets of variables.. i am intrested in software or articles or web sites references or your own experience in such an endeavour if anyone is working on this field of finding the critical components of sucess or failure of a strategy.. the ideal process the way i can picture it is to start by setting manually begin and ending periods of success/ failure samples of the equity curve and try various sets of variables to see what is connected or random to these ''period" and IF something stands out try to writte a predictive code that will switch the strategy on and off this is a point B to point A type of approach that i think is the step number 2(instead of parameter optimization) once a system is generating a satisfactory equity curve in a large data set with a big population of trades.. regards
-
thanks it worked it only took me a bit to realize that i had to plot the the study so that tl_new_self will draw... thanks again
-
hi i have a coding problem in multicharts. is there a way to drawn a trendline at subchart 2 instead of subchart 1 where the price is . what i mean by that is once you create your trendline; NTR = TL_New( BegDate, BegTime, BegTrval, EndDate, EndTime, EndTrval) ; Where BegTrval and EndTrval are values of an indicator the line will plot on subchart 1 . how can i adjust the code so the line will drawn on subchart 2 where the indicator is ? thanks in advance
-
an angle is measured in geometry in terms of distance for both axis in two dimensions.. a..if you consider price as distance then how you quantify time or other resolutions? what you have here in one axis is price and in the other a barnumber.. it is quite unlikely to get what we commonly think as an angle from these 2 dimensions b. what will be your switch that will reset the calculation( point a from whitch you start calculating to point b...and reset to point a and so on). **** regarding slope ***** fixed length is a trap because it is an arbitrary assumption,the market doesnt fit its behaviour in a rolling cycle ask john elhers he knows all about it... almost all his published studies are based on cycles and geometry functions and his " reputable " mesa system on openrange breakout if i am informed well... let us know if you come up with something...
-
Regression Lines in Real Time - Charting Software
andypap replied to friva's topic in Tools of the Trade
there is a drag and drop auto regression channel with various inputs as how to calculate the parallel lines ( atr,standdard dev,stand error etc) for free at hightick.com (mark sandiego is the author if i remember well) . tradestation and multicharts compatible... you only move the centeline once you initially set the indicator to whichever bar in the chart and the LRC adjusts it self... you can repeat this any time you want as new data comes in... -
thanks a lot problem is fixed... regards
-
hi i am trying to build a monthly bar counter in the same manner as a daily bar counter in intraday charts ... value1 = value1 + 1 ; if date <> date[1] then value1 = 1 ; replacing reserved word date with reserved word month wont do since you have to enter the specific yyymmdd format that corensponds with a specific month. month(1080101) will return 1 for january for example. is there a way to build a monthly bar counter ???? it can be of great use in strategies... (for example : if first x trades this month have a negative profit sum then stop trading the rest of the month) regards
-
guys this is a technical problem i am doing my best to describe...it is not an ethical problem... if someone knows how it can be resolved and wants to do it can post a solution... i think that this is only a matter of a certain structure of the array in a coupple of lines.. it can be viewed as a general routine (tutorial) in arrays because the conditions i am describing in the code can be replaced by any condition and the value assigned in the array by any value... the benefit of this structure is isolating specific conditional instances within a more general condition without refering to bars back.... every time the counter = 1 the array slot will be filled with any desired value as long a the first general condition is true.... the counter doesnot have a fixed length so the array is filled in terms of length assymetrically... thus you can get an assymetric n = (array size) moving average for example of highs everytime rsi crosses over 70 ( this is not what i am after anyway) or anything else you can imagine of practical use... so far everything works fine exept the fact that untill in the current run of the general condition , the occurances reach the size of the arrray, the array is bringing forward values from the previous run of the general condition... regards
-
if you remember please post the title i would like to read the thread if this is too difficult to be done the only solution i can think is that it can be included as an ohlc data stream straight from the data source by demand from software like multicharts i will post to their forum and see what they will say about it regards
-
this is the code that i am using vars : cnt(0),avg(0),maxval(0), ii(0) ; arrays ; carr[10](0) ; avg = average(c,1000) ; cnt + cnt + 1 ; if rsi(c,14) crosses over 50 then cnt = 1 ; if c > avg then begin if cnt = 1 then begin for ii = 10 downto 1 begin carr[ii] = carr[ii-1]; end; carr[0] = c ; end ; end ; if c < avg then for ii = 10 downto 1 begin carr[ii] = -1 ; end; maxval = maxlist(carr[0],carr[1],carr[2],carr[3],carr[4], carr[5],carr[6],carr[7],carr[8],carr[9]); here is a scenario of how the code should opperate suppose that while the condition c > average is true there are only 5 instances where cnt = 1 the 5 first slots of the array should hold the closing price of those instances and the remaining slot( array[5] to array[9] the dummy value.. then when c < average the array should reset all its slots to the dummy value then when c > average again the array should start populating again every time cnt = 1 note that the array will populate all the slots with a closing price only if the ocurrances of cnt = 1 are greater than 9 in the current run of the close been greater than the average... sneo (thanks for the input) this is not about periods back... it is an array of 10 insctances of a condition(cnt = 1) within a broader condition(c>avg) that might ocurr or not if the broader condition will switch(c<avg) before the sum of instances of the instance condition(cnt=1) will reach 10 ... i hope this will clarifie everything regards
-
the subject here in this thread is quite simple how can someone compute the price that most and average volume happened within a bar. i think that this particular information will help building averages and other studies more efficienly than using the closing price. these 2 extra data prices are not tottaly arbitrary like the closing price they can be viewed as sequencial (for bar interval) volume weighted data price... does anybody know how one can access such data info because building it from tick charts can be quite stressing for the computer and not everybody has years of tick data in storage... regards
-
well i dont exactly understand what you mean but thatks for the reply anyway i will try to explain further what is the problem i am dealing with. when the close just became greater than the average and this represents the basic condition every time rsi crosses over 50 (second condition)the counter reset to 1 and the value of the close is stored in array[0]... the array has 20 elements... in the following senario the first time rsi crosses over 50 after c > average array slot[0] stores the close of that bar and the rest of the 19 slots are carying over the closing value of the bars where counter = 1 from the previous time that the close > average. what i want to do is every time the first condition becomes true to reset all the array values to a dummy value (-1) and then every time counter = 1 start filling the array with the close... note i dont want to resize the array(dynamic array). then i want to get the max value of the array slots. note that in case of the scenario aformentioned the array will be array[0] = c; array[1] to array[19] = -1 max value of array = array[0] when the second ocurrance that counter = 1 will happen while c > average then the array will be array[0] = c of first ocuurance array[1] = c of second occurance array[2] to array[19] = - 1 if array[0] > array[1] then max value of array = array[0] else array[1] my problem is coding the above efficiently the way a pro would have done it so the code will be precize and 100% accurate my efforts so far require a lot of lines and i am 80% accurate.. for example i am using the maxlist to get tha max value by naming all the array slots one by one value1 = maxlist(array[0],array[1] ......array[19]) ; regards
-
hi i am dealing with a coding problem i dont know how to resolve efficiently what i want to do is first: to reset all the values of the array before last time a condition was true to the minimum value. then : to get the maximum value of the array. i am using a for down to loop to fill the array whitch has a fixed length and two conditions. counter = counter + 1 ; if rsi(c,14) crosses over 50 then counter = 0 ; if c > average(c,1000) then if counter = 1 then begin for nn = 20 down to 1 begin array[nn] = array[nn-1] ; end; array[nn] = c ; end; now the array has twenty elements some of whitch are stored from the previous time that the c was greater than the average when the counter equals 1 . what i mean by that is the if the array evey time the condition (c> average) will start it will carry over values from the previous time that the condition was true. how can i sort this mess out ? regards
-
General Discussion for Components in Strategies
andypap replied to andypap's topic in Technical Analysis
there is no need to match an effort to increase productivy and save your self from a lot of headaches with unequal and useless irony and hostility unless everything is quite obvious to everybody at first glance... i was not aware of what i have proposed till recently and i realized it by accident... i was making a " volatility indicator " to face something that i was involved with at the time and then i plotted it in spx daily just to see and voila the market NEVER show a mesurment above a particular threshold unless it was diving down ("" buy the creepers sell the leapers "") NOT ONLY THAT BYT WHEN THE MARKET WAS EXCEEDING THAT THRESHOLD AT VERY EARLY STAGE OF THE DOWNTREND. each and every downrend from 1960 till today. so i realized that i have a solid threshold for risk measurment i have adjusted my strategies (size,stops targets) and most of all the boldness of the decision making part of the strategy and enhanced the results by multiples ... of course this subject is in the realm of intersest of swing strategie devellopment and the structure of it should simulate a fund behaviour for optimal results and definately has no impact at all to scalping strategies or intraday strategies.. but just this fact can lead to another deduction about markets(another topic for discussion). ///////////////////////////////////// two eyes,one brain what is the level of awareness at all times particulary when you are buried in details and specifications and small tasks...how easily can you drift out of context.. my experience is that you can wonder in a maze for endless time overlooking the most essential so no secret at all only a reminder whitch i wish that someone would have proposed to me to keep in perspective.... there is no harm done here if anyone is buiding a swing or intraday strategie can exchange info and guidelines with others.. i am only trying to initiate the conversation... if people start posting intersting views and structures i dont mind listening more and speaking less regards -
General Discussion for Components in Strategies
andypap replied to andypap's topic in Technical Analysis
strategy devellopment hass many many pitfalls and definetely it is not an easy task.. further more from the perspective of the average trader it is a long term commitment regardless of the reasons that intiated that decision... one person on his own time and resources will face a very long and hard to travell road with the different aspects of creating and refining a strategy... if someone has advanced programming skills and math education and enough trading experience and works with a group of others of various experties required then he has the luxury to show up and dazzle us with his super technologie and brilliant algorithms... but the question in hand is totally different from this approach.. although required most essential criteria apart from enry,target,stop is adapting in "current market conditions" thus achiving normalized equity graph through time and having an accurately estimated execution in the backtester that will limit slippage in reality ,can that be done with relatively advance programming , standard technical knowledge and simulation of a real trading risk managment i think that this description fits most of the members of this forum that want to create their own strategies... but maybe i am wrong maybe everybody has a strategy trading his account with no judgment calls and i am the only one in the dark or maybe people are throwning the white towel after a few frustrations and have made up their mind allready that this is too complicated to achive ... summation of the above is that people that have a strategie, not heard about one ,can ensure the rest of us that this is not a pipe dream and lay general guidelines without revealing their intellectual propertie , while others like me who think that they are close can exchange info revealing as much as they feel like... here is something that i have strong evidence regarding the snp 500 and can be a topic of discussion. an up market is very different than a down market so one should design these two separetely... regards -
this thread has a particular purpose for anyone in the forum intrested. to open a general technical discusion about 100 % autommated strategies... no judgment calls at all.... actually it is about what to keep , what to throw whitch might interest a lot of the participants that try to design their own fully autommated strategies.. let me introduce myself. i am a trader and analyst for 15 years and i have average or even below avarege easy langauge programming skills. i persist though so i have managed to design a lot of my own concept indicators and strategies as well as standard concepts like comressed timeframes indicators,dynamic adaptive to market conditions averages divergence studies, forward channel like price projections, autotrendlines ,statistical studies etc. i have posted many times questions in this forum trying to find programming solutions to given problem that i was facing... i am working on my own so this process was and will be consumming a lot of time and effort so i have dedicated all my free time to it . i will try to be as brief as possible to present my bias and in order to set a frame in whitch the discussion will be conducted lets start with the data first.. price axis, timeaxis ,O,C,H,L,V and timeframes. to start the controversy i suggest that timeaxis , O , C , and timeframes are holding no information of value at all... the market is a series of transactions each at a given price and the only info of any value primarily is what price , how much . the open/close of one timeframe is one transaction whithin a bar of another timeframe.. the time stamp of a ceratin transaction is totally arbitrary and depends on many factors.. to decide that a bar will form every n time doesnt represent at all the the flow of the market whitch is the flow of transactions... so what we are left with is high , low , and volume. more specificlly derivatives of high and low like support/resictance ,range etc. apart than just suggesting it i also testify that i have backtested a lot of startegies based on various indicators that derive from calculation relative to the close,multitimefame,divergence and so on and they were all having random characteristicts. on the other hand strategies(es,forex) based on support resictance , range and other derivatives of high and low are performing very consistantly..and some almost fractal and i am talking about a sample 0f 50000 thousand trades withing 10 years over 20 trades per day the equity curve of those strategies doesnt even flicker whether it is an upmarket,downmarket,sideways market they kind of match the nature of the market exept one "little" problem....execution(spread between backtested hypothetical executions and real live execution) buy that is a topic i would like to discuss later on in this thread if anybody has a different bias or has something to implement in this first topic of this thread is welcomed. the reason why i would prefer this discussion to progress from topic to topic is because it can become quite chaotic if the conversation drifts out of context. regards
-
excuse me for intercepting this discussion what about the most forgotten principle evaluated these days that price is orbiting arround the present consesus of value.. if that is true whitch i think it is the only mass is volume deposited as a concrete foundation arround value with little or no fluctuation at all the rest of the activity no matter how intense it might look is nothing but a passing storm... and another thought to exchange opinions what is closer to the nature of the markets math and physics or plain linear geometry regards
-
this is something i have allready guessed... i design my own strategies in easy language and to be honest with you its quite tough to raise the bar above 70% accuracy and maintaining a good profit factor so i felt a bit intimmidated.. i thought that if so many products claim such huge success maybe i am missing something here and have to study more...find a different edge but then again the advertising of these products is so saturated with hard sales techniques and promises that make me very suspicious.. thats why i wanted to see if somebody has a personal experience to testify... if half of what they claim is true i will quit now and get one of those robots and forget about easy langauge and all the headaches it is bringing to my life regards
-
does anybody have any experience with forex robots.. they claim 90 percent winners / loosers backtested over 10 years of tick data and some claim that they run 1 year or more on real accounts with similar results.. their price is below 100$ one time purchase the only thing they let you see is an equity curve and they bash your head with aggressive sales pitch promising a very fat bank account most of them are metatrader pluggins.. what is the catch here? do they make their money on spread ? are they reliable? regards
-
The markets turn for only one reason - a change in the balance/imbalance in the order flow. what is this a new religion? what about value why is volume not expressing it self in amazon sending it to 3 gadzillion dollars post the time and day stamp of those charts displaying intense volume activity and most likely (i dont want to be arrogant) i will explain to you why price has changed direction there.. volume only represents the perception of value beween market participants and value is inevidably corelated with price. what do you do first when you by a house or a car? do you take a look at the price tag or do you ask how many where sold the last hour. as for technical analysis dying because of computers.. check those pics and tell me what you think... can you code a system that does that... since you obviously cant then you are prone to start thinking entering the room from the back door regards Liquidity Rebate Orders Some of these same “algo” (short for algorithm) institutional traders also receive payment for posting orders, further encouraging huge trading volumes and market volatility. The broker dealer doesn’t even have to make money on the trade itself, and if the order is filled, the broker dealer receives a rebate. These rebates are driving up the price that other institutional investors pay. Many other predatory practices are described in more detail in this Themis Trading White Paper. Bank of America/Merrill Lynch needs to recoup its losses so they just rolled out new options order algorithms as mentioned in Wall Street and Technology. Flash Orders Flash orders are orders that are sent briefly to a limited group of traders prior to displaying them to the public. The high speed computers described above take advantage of this millisecond period and allow their members to trade ahead of those orders and at better prices. NASDAQ and BATS (another equity market in addition to the NYSE and NASDAQ) argue that they’re trying to improve the orders. Even if the orders aren’t improved, they claim that they’re already reviewed by so many dark pools that there really isn’t an advantage to sending flash orders. So, if there isn’t an advantage, then why are they sending them? Flash orders aren’t regulated by the Securities and Exchange Commission (SEC), but according to The Washington Post, New York Senator Chuck Schumer wants to introduce legislation to prohibit their use if the SEC doesn’t take action soon. Some flash orders are sent to Dark Pools, another threat against individual investors.
-
hi guys this code /// vars : xx(0),flag(0); arrays : darr[20](0); if BarNumber data2 > barnumber data2 [1] then for xx = 0 to 19 begin darr[xx] = close data2[xx] ; end ; if darr[0] <> darr[0][2] then flag = darr[0] else flag = 0 ; plot1(flag); ///// is storing the closing values from a range compression (data2) symbol the next bar on a time compression (data1) symbol.. is there a way to take the average N or the xaverage N value from data2 only when the flag variable > 0 ? regards
-
hi i have two questions about arrays : 1: how can i populate an array like myarr[10](0) ; untill a certain condition is true... if arr[n] element falls out of the condition true statment then arr[n] and all the array elements>n should be populated with a zero value.. example cond = c > average (c,50) ; value1 = highestfc(h,mylen) /// mylen is a counter that increments by 1 once cond = true/// if h > h[1] then begin for ii = 10 down to 1 begin myarr[ii] = myarr[ii-1]; end; myarr[ii] = value1 ; end; note that this array will hold 10 elements if from the 5 element down to the 10 element the c < average (c,50) then all this elements should hold a value of zero.... 2 : once the array is populated then how can i get the maxlist off all the elements stored in the array ? regards