Squawk / 22 July 2016 at 8:21 GMT
Hypothesis Testing
United Kingdom
Quant the article commentary is getting a bit long now so for the bare bones algo we also mentioned here is day 2 of the trade set up.
Remember you wanted an absolute basic function to measure for Man/Machine overview. Yesterday the eurgbp suggestion for a max two trades from a given illiquid time point value which needs to be reset (turned on and off as you asked) has worked today so in theory I would have turned the machine off now.
Original concept in commentary here;
https://www.tradingfloor.com/posts/may-the-force-be-with-your-algo-7928840
1y
quant quant
Thank you. I will go through this artice (and others) again on the weekend. The format of the articles, comments and squawks makes it easy to get lost.
1y
quant quant
Now. Something for the weekend.

MAN vs MACHINE

(as fxtime called it - some kind of modern turtle experiment. For the moment strictly intraday.)

So there is nice way for first look at trader performance.

And it definitely not resticted to algo trading ;-)

######

Part #1 of the Rule Set.

You trade for 20 consecutive days.

You have a 'notional' account size of 15K.

Your max. postion size is 2 lots at any time. You trade max. 2 different markets in these 20 days.

You can trade 24h, but no overnight positions allowed. You are flat 5pm NY.

When you are down 1500 USD from the initial account (account < 13.5K) you are OUT.

When you are down 1000 USD on the day you stop trading for this day.

Your target is 3000 USD P&L (account > 18K).

######

1/2 ...
1y
quant quant
2/2 (MAN vs MACHINE)

Some additional techicals:

Lotsize/Markets: This comes from Futures Trading. For the lotsizes if using CFD/Spot/etc use some reasonable equivalent, try to emulate the Future contract exposure. For Spot Forex use 1 LOT = 100K. (And trade always full lots.)

Its about playing the game, not gaming the rules !

The idea behind this ruleset: These assessments rules are made to keep a trader engaged. It gives at least a minimal required sample size to show edge and risk management. You cannot bet on homeruns or sit through drawdowns in this framework.

And to be clear: The rules look simple, but beware ;-)

So start up your excel sheet and calulator, or your development system.
And have a nice weekend.
1y
fxtime fxtime
LOL...clear and concise :-)
1y
fxtime fxtime
As I am in the UK I will use a notional £15k so as not to cause confusion with your $15k....am I to use the last 20 days of trade history or start from today and record LIVE ?
Presume you will be doing the same?
1y
fxtime fxtime
Oops I have just realised you've started another thread so I will utilise those.
However over the last 20 trading days the FTSE cash market has exceeded its open range by a minimum 9pts and triggered 22 successful trades to bank that move and one duff trade incurring a 32pt loss. Normal lot size equivalent would in this instance be £20pp (1 futs lot equates to £10) max drawdown was 38pts
gross profit would thus be 9 x 22 = 198pts x £20pp = £3960 gross profit
One loss at 32pts x £20pp = £640 loss

Net profit = £3960 - £640 = £3320

Max drawdown was 38 x £20pp = £760

Opening bal of £15k would after 20 trading days on this one trade scenario would have exceeded your target revenue and balance.
1y
fxtime fxtime
So for fun lets reset and trade LIVE as it were from Monday ?
We can both post on your other thread?
1y
quant quant
Please. Don't take this to serious.

Just to stir things up a little bit. :-)

Even these simple rules are hard to acomplish for Machine AND Man.

It is just a basic framework.

And for the audience here: can simple algos with simple rules get you somewhere?

Or can they add reliable your reliable performance. (I think this is one of your main themes)

I am just helping out with a little context. How to mesure performance.
1y
fxtime fxtime
Definitely agree....ALL machine systems have to be constantly ''optimised'' for functionality....usually I reset each and everyday !!
I think you've caused a strong interest and hopefully others will join in too.
1y
fxtime fxtime
Measuring performance is a topic I have mentioned many times elsewhere....whether it is beating the mainstream index, beating the risk free rate or simple percentile profits. Another measure I ramble on about is consistency and the frequentist overview of success rate, drawdown, net profits after costs, max consecutive wins and losses and minimums but I suspect quite a few here do not record such data so it will be great to see what comes of your posting. :-)
1y
quant quant
Few weeks ago I started my own discretionary daytrading experiment. (as you can probably guess from my posts). Some stats about day structure, signals and algos only for broader market conditions. Everything else pure manually. Heavy based on auction market theory. (Orderbook/BuySellClassification, VolumeProfile and "Context") All the stuff thats so hard to put in algos. At least with the tools avail. for me.

So maybe I post some stats. (I am a bit reluctant at the moment to do this)

Because you know: if your algo looks bad, you stop it and you still look smart ...
1y
fxtime fxtime
LOL....I know what you mean....to be honest I have been in this business too long to care LOL.
1y
quant quant
FTSE. Cash. I have no reliable real time datafeed for this. And just for fun or experiment I don't want to trade future contracts this heavy. And with the derivates: Especially this market gives me real bad experience for most of the time. I run all my algos here just for fun and on demo. When you posted your "Lady Luck" article I created a quick one off for the MT4 platform and started it on 3 CFD demo feeds. Incredible random results. I seems nobody shows opening or closing quotes with any connection to the real market. Maybe for intraday breakouts the quotes look more sane.. Should give it a new try.
1y
fxtime fxtime
I posted an amended Lady Luck strategy somewhere to show you can improve on the random luck scenario and market skew of 3.1% by using a RSI 5 on an EoD basis using the USA closing bell as your time stamp point. Never trade a long only trade if the shortened RSI 5 is showing a value at close of 20 -51 or above 90....so enter a lady luck trade when values show 1 - 19 and 51 to 90 !
I just wanted to show random luck can be profitable but only just over a year because the markets are 53.1% per year sideways to rising intraday to a minimum move of 2.5pts !
1y
quant quant
I have seen this. Nice. From trading every day for the initial concept, how many trades pa now with the new one?

But the execution problems killed it for me for the moment. In any incarnation. I have already a workaround ready for this. But not implemented yet.
1y
quant quant
GBP. Breakouts.

In an previous post you showed your system for it. Do you still use this? Do you have updated stats?

I traded a perfectly stupid one (not aware of any market conditions or adaptive to anything. Just binary: On/Off. it trades the breakout levels or not).

It stopped it trading, because I thought the trade management should be better. But the proposed changes exposed a (still unresoved) platforn bug ;-(

But a start for my algo part:

I will put it to work an Monday again. At least on demo. Bad as it is. Even if it just to put some concepts on the discussion list and get some stats. (so vacation is over for this one)
1y
fxtime fxtime
The gbpusd (cable) price break out trade is still used but remember it comes in three parts and I think I posted two and both give robust results....I will post the dat set for you in a few mins.
1y
quant quant
Something organisational:

Is it possible for you or your Saxo editor to post a list with links of this years main articles?
1y
fxtime fxtime
You aren't the first to ask that question and the editors here suggest people simply scroll back through everything but I post on average two trades and one article with follow ups per week so there is a lot to scroll through. Once I send the article I have no copy as it is written on the Tradingfloor software not mine unfortunately. A library reference point of any traders posts would be useful imho.
1y
quant quant
the scrolling experience is somewhat limited ;-)
1y
quant quant
Breakouts. Today. A great day for it.

But as trading usually goes: next week we get a new whipsaw experience for every breakout trade.
1y
fxtime fxtime
Cable price break out strategy 1 ....this is the current year (2106) outcomes todate. Real trades. Bare results with no enhancements....
149 successful trades earning a minimum 10 pips NET = 1490pips total
11 unsuccessful trades ...total loss = 436pips lost
Average win obviously as this is a base line minimum 10pips
average loss = 39.64pips
Fixed stop is 45pips
Net profit 1490 - 436 = 1054pips banked for this strategy using spot market prices. NOTE you can apply your own stake size to this as a net profit value in $ terms.
More than one trade can occur in any one day but the the system will not permit two trades to run at any one time for this scenario. Drawdown is obviously nicely constrained to the stop parameter.
1y
fxtime fxtime
LOL....some types of price break out strategies can accommodate whipsaws but usually I work in a single direction.
1y
fxtime fxtime
The adapted Lady Luck eliminates virtually all losses ! It isn't perfect ...nothing is imho but worth doing none the less....you can simply do the trade at the closing bell manually. I cam across the market bias when studying 30yr T Bonds and then applied to SP500 !
1y
zefy zefy
Apologies for interrupting, but regarding list of links to main articles I have found useful to type following to Google search:

site:www.tradingfloor.com fxtime article

It gives you a list of fxtime articles from www.tradingfloor.com in Google search results. Order of articles in search results seem quite random but scrolling 4-5 Google results pages instead of full history of fxtime in TF is more straighforward imho :D.
1y
fxtime fxtime
Price break out #2...this is a smaller pip gain and the stop is max 40pips or whatever the trading range of the triggering candle happens to be if smaller than the 40pips.
These type of trades are recorded on rolling 90 trade events but I can easilly get the year todate data for you.
90 trades opened
88 trades were successful at max stake size but only +5 net points gained = +440pips banked over the last rolling 90 trade days...only 1 trade per day
2 losses for a total of 90pips
NET result 440 - 90 = +350pips booked NET.

This ONLY relates to the spot cable. I also trade the eurusd like this along with the Loon and Yen but the eurgbp is strictly traded via ratios as per prior algo posting.

Zefy thankyou :-)
1y
fxtime fxtime
correction stop is max 45pips
1y
fxtime fxtime
Sheesh I clearly post too much doing that search !! Thankfully I don't stammer LOL. Seems I should cut down on the updates etc.
1y
fxtime fxtime
Cable price break out is actually a protocol point really and i haven't as yet posted an article on points of interest/points of control and protocol points...suffice to say i will at some point and it is income generative.
1y
fxtime fxtime
Forgot to say on price breakout #1
min consecutive wins = 1
max consecutive wins = 38
min consecutive losses = 1
max consecutive losses = 1
So any losing trade should imply you have a higher confidence in the subsequent trade and increase your stake size slightly to form a type of cost reduction management akin to option players.
1y
quant quant
"Lady Luck" / Overnight Trades

I was trading overnight systems (strictly Close to Open) in the US index futures a while ago. Filtering was absolutely necessary. But even with the strategy degraded over time. Still profitable overall, but sideways for very long time periods again and again.

Then I looked at the European Indices. I wrote something in R to graph the Overnight performance vs the Day performance. Very astonishing results. (There is a reason that I want always the see the development over time. The pure percentage numbers are fine, but shows not the trading reality. Maybe I can post some examples)
1y
quant quant
The conclusion of this study was:

For this kind of trade you would never look at the US markets again..

BUT and IF.

I would not bet the farm on it. For now. There are still a lot of issues with this study:

- firstmost my R programming skills
- and I still question the underlying data quality. (Had different sources YAHOO, my own collections and now Quandl. )

- and -- this is a statistical observation, nor a trading model.

So it has to be proved with commisions, spreads, ....

And as ever: some obvervations are not tradeable or exploitable (at least with normal market access)
1y
quant quant
@ zefy

Thank you. Same approach here.

With the new IoT (Internet of Things) I will start using GOOGLE even in my kitchen ;-)
1y
fxtime fxtime
I use ID data via Ensign and manually do the trades....the critical point is (as you say) timing....genesis technologies use a close value some 15mins after Ensign or Saxo and provide totally different results. Spreads are 0.25 fixed on emini as this is the tightest spread but I reference the trade from the cash market. As a matter of course I place a buy not at market price but 0.25 below (old habits from my dealing days and option days) and it is very rare not to be filled. I will download a spreadsheet for you for 2016 data.
1y
fxtime fxtime
BTW NEVER bet the farm on anything....trade small and stay solvent. LOL.
1y
quant quant
Who wants to own a farm anyway ;-)
1y
quant quant
SPY Overnight vs Day
1y
quant quant
GDAXI Overnight vs Day
1y
fxtime fxtime
Straightforward Lady Luck for 2016 on SP500 trading cash results for now are negative after costs and spread
107 successful trades of +2 = 214pts banked
26 losses with a 9pt stop = 234pts lost
But using the RSI 5 suggestion we have 11 of those losses avoided = 135pts lost
net we gain a profit of +79pts
Now working from the cash charts but trading the emini as a tighter spread we get 109 successful trades = +218pts less 135pts losses = +81pts
Remember this is supposed to be a basic starting point to challenge traders belief in various systems available in the general marketplace. I even described an experiment that occurred with chimpanzees as a reference measure !!
1y
fxtime fxtime
Yeah those charts tell how grim things can get but you can skew trades even these ones to give you a very slight edge. It is like option traders thinking to open a long trade could by BUYING a CALL to open but often more experienced traders would SELL a PUT to open and adjust the strike price accordingly to ensure they have the same directional trade but they pay less for the tarde and get a better breakeven.
1y
fxtime fxtime
I do agree the Lady Luck trade is weak but it wasn't meant to be a crux trade structure....they came in later articles. The point was to challenge underlying beliefs that you can only trade using fibbo's or EW and more importantly make traders ask how the propr desks and real major trading desks trade?
1y
quant quant
And don't forget this P&G stuff: Head and Shoulders ;-)
1y
fxtime fxtime
LOL :-)
1y
quant quant
BTW.

For TA Andreas Clenow is your man. ( hegde fund guru and author here on the site.)

He wrote an hilarious article on his sit; "Why technical analysis is shunned". And initiated a heated discussion in his comment area.

In his artice is my all time favourite chart picture.

(Don't want to link other site - not sure about the policy here - but google for " andreas clenow following the trend technical analysis")
1y
fxtime fxtime
LOL....I have seen the article on his own website and I see he too has now done an article on the chimps study on site. Late Aug I will actually be near his offices....I am half tempted to try and visit him !
1y
quant quant
But other tribe: His chimps are trading equity.
1y
fxtime fxtime
haha very good :-)

My article next week is really about the ''in-efficient market pricing'' and I suspect there will be more comments about how wrong I am with that..... than the commentary received this week! Such is our embedded beliefs about the marketplace. :-(
1y
quant quant
Was a nice day in the squawk. Thank you. Have a nice weekend.

And for my "cable trader" part:

Missed the big one - twice. So much for excellent trading execution makes you succeed.

No P&L was harmed. Or touched. ;-)
1y
fxtime fxtime
Have a good weekend too :-)
Virtually all my trades have been covered. Only positional trades remain on SP and FTSE.
1y
Qi2 Qi2
Hi Quant & fxtime, I love this idea, will find the new thread, but see no reason why I can't try this too :)
1y
fxtime fxtime
Good news Qi2....you done well on the competition before if I remember correctly :-)
1y
Qi2 Qi2
Haha, yes fxtime I have, but I have also been hit a few times, I seem to remember that more clearly, I was beign egged on by payback at the time and I was 300 lots of DJI (now US30 CFD) and so I got smoked!
1y
Qi2 Qi2
sorry, had a "300 lot" position
1y
fxtime fxtime
Ah sorry I didn't realise you were mailed by the DJI...I recall you were very high up in the ranks at some point though.
1y
quant quant
These CFD 'Lot' sizes. Always shocked when i see this inflated numbers. Especially in the morning. ;-)
1y
fxtime fxtime
Quant my first trade scenario using your restrictions has just triggered...as described in our prior commentary...I traded the price break out of the ftse cash 100 mkt at the equivalent 2 lots size so just £20pp (rounded down stake) and stop was only 25pts so well within the drawdown parameter and I have reduced the take profit parameter to enhance the probability value. So covered at +7 only. A meagre £140 booked but costs will bring this down to +£130...Acct balance would now be on your criteria £15130
Chart attached
1y
fxtime fxtime
Update....I will for today only (your restriction on no overnight positions) leave an order to short the same market at 6722 at the same stake size with a stop of 30pts only. Take profit would be 10pts gross.
1y
fxtime fxtime
Update day only range break trade has just been placed 10184 on DAX cash...if filled stop will be 50pts...again £20pp stake size and take profit will only be 10pts gross. Order auto expires at 6pm today.
1y
fxtime fxtime
Yikes I can't screengrab and write up the strategies before they trigger !!
1y
fxtime fxtime
Trade covered +8 gross = £160 profit less £10 costs
Current NET balance on the theoretical account rises to £15,290 and negligible drawdown occurence
1y
fxtime fxtime
New range strategy comes to fore.....preset cable short order day only at spot 1.312100 which sells into support ! 45pip stop this time and will halve trade if in profit at +10 with a 15pip trail stop on remainder :-)
1y
fxtime fxtime
Apologies but I have a meeting now so will leave these for current testing by you.
1y
zefy zefy
Seems I am not getting anywhere today so nothing to report yet.
1y
fxtime fxtime
Just back to my screens and see the cable trade described has just triggered at 1.312100 and the trade was halved for gross +10 and I have manually covered the trade at the same point too as I will simplify these trades to bare bone trades without any silly alterations so +10 overall but costs rose to £15 due to my manual closure and revenue was £200 gross and nets out at £185profit
Zefy how are you doing now?
1y
fxtime fxtime
Sorry I should have said net balance rises to £15290 to £15475 and drawdown what little there was...proved to be well within your criteria.
1y
fxtime fxtime
Had I left the trade to trail then profits would have improved but we looking at basic algo or manual trade structures with DEFINED risk. So far I have been ''lucky''
1y
fxtime fxtime
Just checked my account balance and it appears my daily usdcad trade that I post onsite for net +5 also triggered eg pbo of the 11-12noon BST price fixing point. Again £20pp stop in this instance was stupidly small at 21pips (eg the trigger candle range)....this is NET structured trade eg the target profit is only after costs +5 as that has the highest probability....just look at my prior postings for the usdcad for the trade set up....there are dozens of updates for this one LOL.
Net profit rises to £15475 + £100 = £15575 (after all dealing costs have been applied) note interest charges would also apply and I will calculate at the close and debit the £15575 balance to keep this balance relevant....so far all vanilla and plain :-)
How are you doing Quant/Qi2/Zefy?
1y
quant quant
Thank you for for the very detailed stats..

I will update after NY Session.

To distracting for me intraday.
1y
zefy zefy
I am at 15006€ LOL. Traded only FGBL (not theta-trade) and timing in the morning was horribly off.
1y
fxtime fxtime
Hey a positive return so as far as I am concerned it is a good result. Especially when you consider the risk free rate on a daily basis equates to 41 cents per day so you've beaten your savings deposit rate !!
1y
fxtime fxtime
Update the ftse trade outlined above which had a short order outstanding at 6722 was triggered and completed for a gross +10 as described this morning ...dealing costs were £10 in total (rounded up) and gross profits was £200 therefore only £190 banked taking the account balance to £15,575 plus £190 = £15,765 net. A reasonable day return for a small scale account imho :-)
1y
fxtime fxtime
Final update on this mornings unfilled order....DAX did indeed fall and my 10184 order was filled and +10 was banked as per the plan above.
£200 banked less deal costs of £10 equaling £190 which when added to the £15765 balance above brings me to £15,940
There are no more outstanding orders pending for today so will resume tomorrow if required :-)
1y
Qi2 Qi2
I'm now going to focus on my other system, but had to try the GBPUSD system I have, as its great this time of day and ratio in a few EURUSD
1y
fxtime fxtime
Wow that is seriously brilliant...you beat us all mate. A great call :-)
1y
fxtime fxtime
A lot of trades but clearly profitable even after costs.
1y
Qi2 Qi2
And the spreads are ridiculous, compared to my multi dealer platform
1y
Qi2 Qi2
Here's a Feb 29 to April7 2016 run
1y
fxtime fxtime
If that is real money then simply shut your screens down and retire mate....astonishing trading...remind me to never trade against you in a trading competition LOL.
1y
Qi2 Qi2
Haha, if that was a real account, I would be call it a day! I created a system that trades EUR/SUD and EURGBP against the GBPUSD. As I mention in your other strings, it needs revisions, as it's filtering is EMA reliant. It is quite successful on a multi dealer network, we have been running live since early May, and it's reasonably successful :) I have been granted a funded account, which mirrors my account, minimum position though is 1,000,000 per trade, it's causing me some anxiety :(
1y
fxtime fxtime
LOL the 1m accts do cause ''anxiety LOL but strangely at 10m > they don't as you become more focused on the pure numbers although I couldn't trade the volume of trades you do for your clientele as mine would have a heart attack seeing the sheer number of trades and demand I slow down. You have clients who are totally risk aware and have no aversion to it.....also the larger accounts as you've probably found already are harder to grow as you will start to diversify your holdings as the fund balance constantly grows and then there is the age old issue of hedge-hogging.
1y
Qi2 Qi2
This is what the London based company is teaching me and also why the live algo only shows pips, as you will see from earlier postings. Also why they use mirror account (2 millisecond execution-LD4) so that I will learn to trade pips, not value.
1y
fxtime fxtime
Seems prudent and the speed of the mirror acct execution is fantastic !
1y
fxtime fxtime
You can also set the start of the trade day as log normal (zero) which resets all algo's and permits a re-synch of timed parameters etc and allows for adjustments to real world events too.
1y
quant quant
Hi Qi2,
Hi fxtime.

At the moment it seems the algo traders have all the fun.

Lets see what happens in forward testing the remaining days.

There would be ton of question from my side, but for the next couple of week I think I will go read-only in this part of the discussions thread. This almost "no losing trades" scenario is too far fetched for me at the moment.

Just one last question:

How would you quantify the difference between this stellar demo stats and the mentioned "reasonable successful" live system?
1y
fxtime fxtime
Approx 99.9% of all reported trades I give are quoted in pips/points earned only Qi2
1y
fxtime fxtime
LOL my system doesn't have stellar returns sadly....the trades I have offered are the exact same I have been posting over the last six months so my trades are ''robust'' in the parlance of statisticians. However you are right to ask if a demo acct is utilised differently than a real live account....initially yes they are different as risks are taken in a demo acct that aren't in real life. I am perfectly happy to continue the project...you can see from the posts of orders placed in the morning and posted as such that didn't trigger until this afternoon so this is about real time posting you can get imho.
Qi2 is operating in a HFT manner and the only way to go for that is via an algo. The law of large numbers will show if a trade structure works and you commit to it in large numbers then on balance a working trade model will earn. The stops and spread are horrifying to me but backtesting (currently) supports the data he has posted.
1y
fxtime fxtime
Perhaps the best way to overview is simply let time pass and we see over the next two weeks or one month on how consistent the performance happens to be? How about I repeat the same trades I have done today each and every subsequent day and see how just these trades posted today perform? The original algo for eurgbp wasn't included in my trade data nor were any normal scalp trades or position trades so as a core trade plan and result over-view it should be conclusive?
Qi2 might not be able to post everyday though as sometimes releasing fiscal returns can prove a hinderance as other factors can interfere eg other TF readers may trade the same function without fully understanding the trade set up and that can become a confusing issue on site. Nor do we want to give every HFT trade scenario away....remember Qi2 has worked hard to build his trade model and would wish to recover costs for all his hard labour(s).
1y
marran marran
fxtime the system we use with the scalp on the usdcad what stop loss are you using?
1y
fxtime fxtime
MAX 45pip but if we have a triggering candle that is less than that as a range then use the lesser value plus spread as your stop. So if the triggering candle range is say 30pips and you open a trade 2pips beyond the range and your trade ticket shows a spread of say 2 pips your stop would be 34pips. Hope this makes sense? Often for the stops though my systems will adjust the orders once triggered and the stop effectively becomes a contra order at double size....thus if 100k long the stop will be 200k short as this reduces costs and turns the trade if need be and still results in a 100k short trade as the other 100k cancels the original trade. I always look for costs reduction and the fastest fill type orders fwiw.
1y
kom75 kom75
nice contest, and impressive results from Qi2. Interesting to see how this contest will finish. Btw. Fxtime for usdchf do you use 9-10 candle as for euro and cable, and aim for 7 pips?
1y
fxtime fxtime
kom I often use 7am - 9am trading range for usdchf ! The 9 - 10 range has a lower confidence factor and really you can only get a solid +5 net in that time frame and the 11 - 12 range :-)
1y
kom75 kom75
Thanks, I ordered ea based on this strategy, so will see results next week. so far i think usdcad is the most reliable. could you recommend any backtesting tool?
1y
fxtime fxtime
Each market reacts to a) it's own economic data release timings and B) international price fixing points such as the 11/12noon time slot.
As for backtesting my Ensign has three means of back tests...
#general tick back test from todays date to what ever the earliest date you want and then operates the same function from the earliest date to the current date. That way you can determine running both ways in historic data if your price breakout strategy is robust as you have twice as much trade scenarios to work with.
#traditional tick and price and vol testing in one direction starting at the earliest date point you set
#time stamp testing eg you not only set the earliest date for the testing but at what time you start the test period each trading day.
These three scenarios also permit correlation and non correlation studies along with any criteria you wish to apply whether it is formulaic eg standard dev or null hypothesis to charting parameters and averages.
1y
fxtime fxtime
Back testing and elimination of curve to fit issues is a huge topic mate. It can be done manually using Bayesian models....I think the old style maths books are still available even now....Kruschke used to be the foremost author on stats models and Bayesian data fwiw.
1y
kom75 kom75
Thanks, I will look at it
1y
fxtime fxtime
Just remeber the usdcad works solely due to a price fixing reference point and the market is moving from illiquid to a liquid state hence the robustness of the trade. The eurgbp overnight trade I posted somewhere here works in an illiquid state as your reference point and relies on much the same event that at some point it will be influenced by a liquid market adjustment (Asia) or Europe coming back on stream and has the same reliability !
1y
kom75 kom75
Cheers mate, I'll give a try
1y
fxtime fxtime
kom does your EA permit the 6 line algo on the eurusd/gbpusd=eurgbp trade ratios as you would be profit by now? The programme seems ridiculously simple but the programming for the trigger and stops takes time to set up correctly so make sure you back test on each individual line.
1y
kom75 kom75
no, it's simply ea based on hourly candle brekout. That what you mention will be my next order:)
1y
marran marran
hi fxtime if you trade the usdchf is the close of the 11-12 candle the best time to trade? same as usdcad?
1y
fxtime fxtime
Yes Marran use the same as usdcad but today is FOMC day so I haven't done the trade.
1y
marran marran
fxtime, i have been thinking of using the scalp system for the nzdusd, but im not convinced of the liquidity what do you think and what time (UK) would you look at?
1y
fxtime fxtime
Interesting.....I don't trade the kiwi but I will have a look and get back to you :-)
1y
marran marran
thanks i appreciate it
1y
fxtime fxtime
Marran I have had a quick look at this pair and so far I don't get a consistent result although that could be my data isn't ''clean'' for this pair so I will put through my main systems overnight and see if I get better results for you.

Disclaimer

The Saxo Bank Group entities each provide execution-only service and access to Tradingfloor.com permitting a person to view and/or use content available on or via the website is not intended to and does not change or expand on this. Such access and use are at all times subject to (i) The Terms of Use; (ii) Full Disclaimer; (iii) The Risk Warning; (iv) the Rules of Engagement and (v) Notices applying to Tradingfloor.com and/or its content in addition (where relevant) to the terms governing the use of hyperlinks on the website of a member of the Saxo Bank Group by which access to Tradingfloor.com is gained. Such content is therefore provided as no more than information. In particular no advice is intended to be provided or to be relied on as provided nor endorsed by any Saxo Bank Group entity; nor is it to be construed as solicitation or an incentive provided to subscribe for or sell or purchase any financial instrument. All trading or investments you make must be pursuant to your own unprompted and informed self-directed decision. As such no Saxo Bank Group entity will have or be liable for any losses that you may sustain as a result of any investment decision made in reliance on information which is available on Tradingfloor.com or as a result of the use of the Tradingfloor.com. Orders given and trades effected are deemed intended to be given or effected for the account of the customer with the Saxo Bank Group entity operating in the jurisdiction in which the customer resides and/or with whom the customer opened and maintains his/her trading account. When trading through Tradingfloor.com your contracting Saxo Bank Group entity will be the counterparty to any trading entered into by you. Tradingfloor.com does not contain (and should not be construed as containing) financial, investment, tax or trading advice or advice of any sort offered, recommended or endorsed by Saxo Bank Group and should not be construed as a record of ourtrading prices, or as an offer, incentive or solicitation for the subscription, sale or purchase in any financial instrument. To the extent that any content is construed as investment research, you must note and accept that the content was not intended to and has not been prepared in accordance with legal requirements designed to promote the independence of investment research and as such, would be considered as a marketing communication under relevant laws. Please read our disclaimers:
- Notification on Non-Independent Invetment Research
- Full disclaimer
- 沪ICP备13028953号-1

Check your inbox for a mail from us to fully activate your profile. No mail? Have us re-send your verification mail