onesmith

Pros don't need built-in icons. Attempting to discredit the OP based on his choice of platform fails. This thread is outside the box. Imaginations already racing toward Genetic Algorithms. Lead or follow. The limits you aspire to resolving don't hinder some of us.

Tradestation is awesome. I've only experienced one instance (over the course of 18 yrs of using TS) where I failed to complete a project that interested me. The stuff you describe as being beyond the capabilites of TS are relatively simple studies that can be achieved in TS. ... (by professionals).

Here's the steps that will refresh the areas that could be corrupted. Database. Download Scheduler settings. Single instances of Chart Objects. 1) View, Chart Analysis Preferences data tab, checkmark show download scheduler request 2) File, Preferences, Desktop desktop tab, checkmark open a new desktop on startup. 3) Close TS and reboot 4) Delete the Cache and TCache sub-folders within the TS\program\ directory 5) Start TS but do not open your old desktops and workspaces. Instead create a new chart. Specify 1 tick interval, 6 months back, for an actively traded symbol like @ES because it's a huge ammount of data that's big enough to trigger a download scheduler event. This first time (only on this first event) .. you will see an auxillary popup window where you must specify the data be downloaded ASAP and checkmark a box that says Set Default. The inability to set this directly via step 1 above makes it difficult to get this set correctly. 6) Close TS and reboot. 7) Manually backup everything by copying the TS directory and all it's files to an external drive. 8) Open TS and open your old workspaces. They have not been refreshed but you just created a backup. It is unlikely there's anything in these objects that caused your problems. If a few charts are corrupted then delete those charts, save the workspaces, exit, reboot and create a backup. If the freezing problems reoccur delete the TS directory and copy your backup to that location. Worst case, uninstall, then reinstall, then immediately create a backup thus lessening severity of future worst case.

I compared the results of the TS covar function to the output of an excel spreadsheet and also an online covariance calculator. Results of these 3 sources are identical.

The primary advantage is readability facilates understanding how it works. It's efficiencies such as declaring squareroot(2*g) only become significant if the concept is viable and you have a way to exploit it.

var: tic(minmove/pricescale), pt(1 point); plot1(tic,"tic"); plot2(pt,"pt");

Do you have the exact specification in a succinct format? Compare it to this formula which uses a more intuitive logic flow to return the same values as the formula you posted. input: g(.0001); var: k(c), dif(0), avg(0), vel(0), sqrt2g(squareroot(2*g)); if currentbar>1 then begin dif=c-k; avg=k+sqrt2g*dif; vel=vel+g*dif; k=avg+vel; plot1(k,"k"); end;

This code is typical of the beginning step to test an idea. It plots the traditional indepMean next to my understanding of your alternative indepMean. The results don't necessarily prove or disprove anything. They provide information that enable altering the formula before proceeding to the next step. The alternative indepMean is considerably different than the concept of mean. You didn't post a probability function so I used the built-in zProbability. input: indep(c), len(20); var: x(0), indepMean(0), bhsIndepMean(0); indepMean=average(indep,len); bhsIndepMean=0; for x=0 to len-1 begin bhsIndepMean=bhsIndepMean+indep[x]*zprob(indep[x],len); end; bhsIndepMean=bhsIndepMean/len; plot1(indepMean,"indepMean"); plot2(bhsIndepMean,"bhsIndepMean");

Can you characterize the level of success you've had in your personal trading .. (of machine written strats)? I'm curious because I am aware of your skills in this area, and I have a lot of respect for you. That opinion won't change regardless of your answer and this is definitely not an attempt to discredit you. It's difficult for me to imagine sharing successful tools with others but I understand and acknowledge there's differences between being able to conceptualize and write the tool .. verses knowing the secrets a skilled trader would seed the inputs with.

Edges are plentiful on the smallest scale where they preference doing something one way instead of another way. Viable strats include significant edges independent of the sum of these smaller advantages. imagine abstract truth that solves many of the markets riddles. significant edge = known truth; formulate understanding of se in a way "that must" result in significant edge. guarantee by the mathematical certainty built into the formula that it must achieve the goal of exploiting that se.

It's good to write lots of strategies. The more the better.

RR incorporated as Pro Trading Course LLC. so why is RR personally named as a defendant? Can the facts supercede and nullify liability limits? Are the disclaimers worthless?

It's not difficult to track the real OHLC (within the context of rules based on a simulated OHLC). Post the rules for the 'renko supreme system' and I will post the code for the TS EasyLanguage strategy that can backtest it.

The changes I made should satisfy your objective to alternate between long and short. {Renko-Adaptive signal (based on ATR) Written by konkop 07.03.2002. ********************* Attention: set MaxBarsBack (f.e. =30) in this study according with MaxBarsBack value in the Renko_Adaptive indicator for correct step-by-step calculations of the buy/sell levels *********************} input: K(1), Smooth(10); var: Brick(0) ,DN(0), UP(0), BricksUp(0), BricksDn(0); Value1 = AvgTrueRange(Smooth); if currentbar=1 then begin Up = H; Dn = L; Brick = K*(H-L); end else begin if c>up+brick then begin BricksUp = IFF(Brick = 0, 0, Floor((C - Up)/Brick)*Brick); UP = Up + BricksUp; Brick = K*Value1; DN = Up - Brick; BricksDn = 0; end; if C < Dn - Brick then begin BricksDn = IFF(Brick = 0, 0, Floor((Dn - C)/Brick)*Brick); Dn = Dn - BricksDn; Brick = K*Value1; Up = Dn + Brick; BricksUp = 0; end; if up>up[1] then buy this bar C else if dn<dn[1] then sellshort this bar C; end;

Would someone summarize what we learned in lesson 1 besides most indicators don't work?

There isn't any discussion of a trading method within the last 33 posts (from post #121 through #153). I did not look further back but I recall the posts prior to that are similar. My post which was deleted for allegedly being off-topic, .. conveyed important information relevant to the topic. I'm interested in discussing the FeltonTrading Method with everyone else on this topic. Particularly with those who have subscribed and have experience with it.

In the first example you gave the value being crossed was 80 for both the buy and sell thresholds so that case might be different than your current case. In the first case if myVar crosses below 80 then it must have previously been above 80, so in that case it is unnecessary to determine if it crossed above 80 before crossing below 80. If your current case uses more than one threshold, for instance positive and negative 80. The specific circumstances of which thresholds must be crossed will determine if you need to track multiple events. If your requirements are that 2 events are necessary then this combination becomes a complex event which is commonly tracked by state variables. There's nothing wrong with thinking of them as complex or compound event vars if that makes it easier to visualize, but events normally define relatively simple conditions. A state is different from an event in that on the bar where the state becomes true it's not easy or efficient to look back at previous bars to determine if all the conditions were met. So a state var tracks the events as they occur and sometimes the events combine over time in a way that satisfies the requirements of the state.

The case you described can be simplifed to .. if myVariable crosses below 80 then plot1 or whatever else you might want to do. This is true because your var can't cross below 80 without having previously crossed above 80, except during a brief initialization best handled via another logic gate that wraps everything. If your true requirements are compounded in ways you've not mentioned you need to declare a variable that serves the function of a state machine. For instance if event1 then state1=1; if event2 then state1=2; if event3 then state1=3; Changes to state1 do not have to be linear. Any event can reset state1 to zero, or alternately increase or decrease it for instance: if event5 then state1=state1+9; Multiple different states can be tracked by declaring multiple state vars such as state2. It is imporant to understand the specifics of the secret logic you're trying to implement before it's possible to determine the best answer to your question. You should practice distinguishing the differences between an event and a state.

"Solutions are all that matter." ~ brilliant mentor who brings urgency to every moment combined with an immense ease.

I had a post deleted from this topic (for allegedly being off topic). It named Roger Felton, FeltonTrading, RF or FT six times within it's 105 total words. Discussion with managment confirmed it was not a personal attack. I prefer this be my last post to this subject (but I will adapt to information as the future becomes the present).

The EasyLanguage reserved word PositionProfit can be used to trigger an exit. if PositionProfit > 0 then begin sell next bar market; buytocover next bar market; end;

Everyone is already sufficiently regulated. Courts are always willing to listen to cogent demands. They have the authority to supercede Acts Of Congress. For instance, defining the exceptions that merit nullifying a Congressional grant of Immunity to a broad class of entities.

I restrict my comments about others to things I know are true and I am prepared to prove, .. primarily because it's the right thing to do., ..but also because 'Libel' is a 'Felony' punishable by imprisonment for up to 5 years and a fine of up to $2,500 under 'Criminal Law'. .. besides the obvious 'Civil' penalties.

I didn't say pattern. I can broaden recurring features to recurring themes and recurring logic. I understand those descriptions include the concept of patterns and I'm not being difficult just clarifying a slight difference. Primarily because I do not have enough knowledge of things you mention to understand them or your intent or the direction I would need to follow to exploit them. Given this distinction where I place less emphasis on patterns and more on recurring concepts, can you elaborate on the subjects you mentioned or take a different approach if that changed anything from your perspective? The repetition I'm referring to never ceases. Code that exploits it is vulnerable to error and being tricked by data that coincidentally mimics a recurring feature but wasn't caused by algorithms like the authentic cases. Writing code that captures the recurring aspects of authentic themes, concepts and logic is done in small steps without concern for whether it improves strategy results. I don't mean to repeat but not sure it's been said this way .. fine tuning the results of a strategy that doesn't have a logical advantage or edge as it's basis .. is essentially fitting to non-recurring features of the data. This can be eliminated by not considering improving the results as adding to or more importantly creating the edge that must be there before giving any thought to fine tuning the results. The strategy results do not matter and are likely to still be unprofitable when it's time to look at them for the first time and begin fine tuning the results. At that time, with a thorough understanding of the edge, it's necessary to rewrite the previous attempts at implementing something to exploit the concept, without ever having looked at the test results. It's near impossible to find a profitable edge in it's raw state and near impossible to learn things about an edge while focusing on fitting (data) to a small aspect of an edge. This is backwards and simply more fitting of nothing to the non recurring features of the data.

BHs, fitting to non-recurring features of price is eliminated by targeting recurring features, not by constraining efforts to fit recurring features. Value is determined by the existence of tactical and logical advantages. This is analogous to knowing which recurring feature of price was targeted. I hope you point me toward stronger truth rather than pointing out my obvious semantic and logic breaches, or suggesting I can gain something by being less cryptic.