Its a 24 hour contract
Its a 24 hour contract
But the SP500 is only open cash hours? How come there is no gaps in that chart?
Do I need to enter in manually the time zones or something to ensure it only shows 8.30am to 3.15pm ?
In your post here:
I see that on the SPI chart you need to adjust it to 9.30 am to 4.10pm to get the cash hours.
I think I'm gonna owe you a Coke when this is all over.
60 hrs a week of work + settling + family health issues has meant trading has been on the backburner last weeks. Will be a big change of scenery for me soon which will lead to much more free time, and I can actually post something worthwhile.
I'm not going to waste my time trying to get all the required data for futures + fx and things I now realise I need to pay for these things and I will do when I can justify it. Will just backtest EOD ASX stock from yahoo to begin with.
So where were we.
Its been over a month since my last update. Iíve since quit work (I finish tomorrow) and Iíll be doing some stuff which gives me a lot more free time. Very excited about this as I feel I can actually make some progress and turn myself into the market/systems nerd I need to be.
Just on that last bit Ė systems. I need to remind myself that learning systems is the current goal. Reading CanOZ talk about market profile or trying to work out wtf Jouless + T/H are saying the banter thread is fun and interesting but its not going to teach me how to make money. There will be a time when those discussion will matter and it wont take me 10 minutes to learn what sinner means every time he posts, but its not anytime soon.
I need to concentrate on developing the statistical knowledge and skills required to develop a mechanical and/or automated trading system.
This means learning the ins and outs of data collection, the software programs and measuring success. Itís important to Ďkeep in touchí but I think reading Bloomberg, ASF, commsec, daily news and the odd blog is more than enough to know whatís going in the broader picture. At the end of the day I need to spend my time learning how to make money.
Anyway, I also moved out on Sunday so all the baggage that comes with that is still being sorted. Apparently TPG will take 10-20 business days before my internet is connected which is a let down but thatís just part of life. From Monday onwards I wonít have internet for a little while and then I go to Europe at the end of August so the next few months will be bumpy but post Europe its all hands on deck and no excuses. Life is short I should put the effort in.
From a trading perspective Iím dying to trial a couple of very basic systems relating to up key reversals and pivot points. When I get to this stage Iíll be blogging everything and asking questions about what Iím doing wrong. Iíve also had trouble with Yahoo Data recently and it just doesnít seem to be working so google finance it is for now. I wont be paying for data for a while yet.
So as I sit here on my 2nd last day of work Iím super excited about the future. Unfortunately the next period of time will be choppy. I donít even have a couch or tv yet.
The idea of backtesting sounds really good, how exactly does one do it though? do you need to learn programming code in amibroker?
Make sure you keep us posted OP - what you are doing at the moment is what I hope to do in a couple of months.
Thanks Steve C.
Make no mistake I still feel like I'm an absolute noobie. That being said I've leared a stack over the last year.
Working in an investment team for the last year has allowed me to develop an understanding of how the market as a whole works. Have also developed a good understanding of different types of asset classes, be it debt/equity/etf's/managed funds/overseas investments etc.
I've also learned a great deal readin this forum. I think the trick is knowing who to listen to and being able to decipher through all the text to find things which apply to you. As I said, right now I need to become far more technical and understand the systems side of things. The learning curve will be massive.
With regards to backtesting, again I'm green. Effectively there a number of charting and backtesting software programs out there which allow you to test concepts. Amibroker is one of them. It has been suggested to me that Ninjatrader is a good place to start as it quite easy to test simple systems. This is where I am up to now.
I'd suggest having a looking at a few (and by that I mean all) of the threads in the trading strategies/systems area of this forum. That's my plan.
I note many are looking at the Systems path.
I think this is one of the best things a serious newbie can do.
It will overtime tell you more about what doesnt work than most anything else.
What not to waste your time with and that all systems arent equal.
Youll soon learn that just about anything works in a bull market.
Youll find that short and being out of the market at times are the best positions.
Simple is often best.
Your discretionary trading will improve.
Your Risk and portfolio management will improve.
It wont be long and youll have a light bulb moment.
Sure there is a lot to learn---but then anything worth while takes time.
Dont be to fast in investing in the market and too slow/stingy in investing in your education.
I heartily recommend Tradesim..very helpful owner, David Samborsky of Compuvision. Only $200 and upgradable to more sophisticated versions when youíre ready for that.
Tradesim works very easily with Bullcharts and Metastock. I much prefer Bullcharts because I can import and export lists to Excel ..my main live data scanning workhorse which catches moves as they happen. And BC is a homegrown Aussie product too with excellent customer care.
Bullcharts and consequently Tradesim (linked language) not difficult to program.. spend a few days with it on your European trip. Start off with something very simple such as MA crossover and discover that profitability is not so easy as that.. MA_Xover can even be quite unprofitable in fact, then graduate from there as your coding skills improve. There are some very generous people on the net..learn from their example coding.
I had a very complex system at one stage, but now I use only 3 indicators which draw my attention to potential buys, 2 simple ones I invented, and a >25EMA filter in a bull market that enhances the performance of the other two. Other EMAs work too, but 25EMA works best for my style of trading.
Put a big sign above your trading desk: Donít be greedy; Donít panic...and above all Donít throw your tried and tested system out the window. Tweak it and back test it again by all means, but not in mid trade. Try to keep emotions out of the loop, and treat the profit /loss figures as numbers...not real money..sounds daft, I know, but it works for me.
If in doubt or you find yourself hoping, praying, burning joss sticks at the local Buddhist temple or having nightmares about your portfolio, stand aside. The market will still be there tomorrow and next week. Do practise Risk and Money management.
Good luck mate.
the other is forward worked, under real stress, encountering key moments and unknowns that cannot be synthetically experienced...... and with the (real time) time constraints
these are things to consider........
I have included costs at .08 of 1% each way which is the brokerage I pay.
I have not included slippage as a set %. I have tried to cover that using the calculatable trigger figures for my 3 automated indicators...i.e using algebra in an Excel cell what must ... the price be tomorrow to trigger
auto calculated momentum x
auto calculated short term resistance/support y
25 ema (an entry criterion, not an exit trigger)
...oops I forgot volume
volume on entry must be at least 80% of its 20 day SMA of vol. Of course you cant normally calculate that till later in the day so I use a predicted figure in the morning based on how the code is performing pro rata of the time elapsed in the trading day so far and also the mood of the rest of the market.
..and there's one other indicator ..expanding Bollinger Bands that is sometimes the icing on the cake turbo charger.
I only trade in the top 150-200 liquid companies which I recalculate each weekend. I eliminate also co's that trade on overnight markets such as BHP and RIO that may give you a nasty shock next day by gapping too much.
My entries are at my discretion, my Stop Loss exits auto.
But you're quite right...what's to say my calculated trigger EXIT figures which I place as contingent orders beforehand will be filled perfectly. I don't know how to deal with that...what do you suggest .... % of ATR?
The discretionary bit comes in when my system selects several codes and I decide by eyeballing the live chart patterns and past daily/weekly patterns and daily weekly support/resistance whether I fancy it or not.
My system itself averages about 1.6 trades per day..which could of course mean more or less than that. I have allowed for 20 open trades which would be hopeless to monitor without auto SL orders.
Of course my system is not the Holy Grail, and I may not in reality achieve the same returns, but backtesting gives me the confidence that the chart patterns I am eyeballing may have the odds on their side.
The lifestyle I lead when living In Thailand for half the year and the time zone doesn't lend itself to monitoring the market constantly. When I'm not in Perth, I sometimes don't wake up till 3 hours into the session. NZ is the best spot for trading the Aussie market I reckon... wake up not too early, a leisurely breakfast, a spot of research, market opens at 12 noon NZtime, and close 6pm ..just in time for a sundowner...you lucky Kiwis, such fabulous scenery too... I may go and join my niece's family living there!
Thanks for your help
Last edited by cudderbean; 5th-August-2012 at 07:12 PM.
So the technician from TPG is coming tomorrow to set up the internet for my apartment so as of tomorrow I'm back from the stone age. Currently sitting in the library enjoying having a pretty easy day.
Between organising my Euro trip, switching careers and moving into an apartment I feel like I haven't had much time for anything else. I'm getting through all my tasks though which is good and by the end of this week things will settle down.
I re-read Murphy's technical analysis book and am halfway through qauntitative trading systems by brandy. I also watched the DVD which came with book yesterday and I found it very informative. I mean lets be honest at the end of the day statistics isn't the most exciting topic but there's some little part of me that likes knowing the maths behind the scenes. Also now that I'm taking this on I don't like losing so I know I won't quit until I've given it all I have.
Howard on the off chance you read this, the DVD didn't work on 2 of my DVD players or my laptop, it only seemed to work on my computer. Also I didnt realise you were from Arizona, I did 6 months at ASU and it was defintiely the best 6 months of my life. ASU girls were really um, friendly.
Was interested to here Howard talk about market diffusion at the end of the lecture. From what I gather, having a strong understanding of market breadth and the overall flow of the market is one of the most important things I can do. I know T/H likes his market breadth and Joule's often refers to put/call ratios etc, I think sinner mentioned something about the mclennan index on here recently to. Effectively they are all looking at the same concepts and assist us in keeping our finger on the pulse of the overall market.
It had me thinking and I want to get into the habit of maintaing/keeping an eye on 5-6 of these 'diffusion indicators' for the ASX and S&P. Not so much to develop any mechanical systems, just to ensure I always have a feel of wahts going on.
At this stage I'm thinking:
Advance/Decline (I like the 10 day advance/decline MA oscillator version)
52 week highs/lows
some kind of option indicator.
I'm sure this list will grow and chop and change as I learn more. Any suggestions on others I use would be much appreicated.
On another note, I cant believe I need to blow the dust of my year 1 lecture notes to remind myself about t-tests. feels like forever ago!
Finally have the internet. That was a rough couple of weeks.
Head to Europe on Monday. Very cool/excited.
Have watched Howard Bandy talking at the national conference in Melbourne in 2009 twice now and I'm still confused about walk forward analysis.
I understand that ''in sample'' results mean nothing and its the out of sample results which are important. With walk forward analysis do we optimize after each out of sample period? Then these new parameters are now our system?
My understanding was that if the system performed well on out of sample data it should be traded as is but looking at the walk forward analysis picture in the link below this isn't the case?
Still planning on making all the money in the world.
The more I learn the stupider I feel
Well 2013 I'm giving myself 1 day a week devoted to trading. I'm lucky enough to not have to answer to a boss and I'm giving myself Wednesday off each week in order to learn how become the greatest trader in the world. I figure it shouldn't be too hard. It's actually a welcome change as my normal routine is pretty stressful so breaking it up midweek is a good opportunity. Plus at the end of the day its what I want to do so I may as well make a run at it!
Today it took me most of the day to work out but I have finally got the SPI set up on Ninja trader so I can watch it and play around with it. Also worked out how to set up day and night sessions as well as how to use the session manager and a couple of other neat tricks on NT. Slow and steady at this stage but I think I'll slowly get my head around the software.
As far as goals I recognize I'm light years away from trading real cash. I've set myself the first goal of developing some EOD systems for Aussie stocks and then taking it from there. This blog from here on out will be the tale of my trials and tribulations as I learn to use the NT back testing facilities. Hopefully the things I learn from that will one day lead me down the path of intra day work but only time will tell on that front.
Truth be told I had no reason today to spent so much time worrying about setting up the SPI but it's all a learning process and now I can sleep at night knowing I have it working properly.
So I had to run a few errands this morning so my 'trading day' was cut a little short. Still a productive day though.
This morning I continued to get accustomed with Ninjatrader. I set up a Market Analyser list which is a pretty cool function. I've effectively put all the major indices/currencies in their and I plan on adding to it over time. I think it's a good way of getting a 'snapshot' of what happened overnight etc etc. Unfortunately it's not really useful for me at this stage as I need a life data feed in order to view the data but hey, all in good time!
I also watched a 2 hour video from Radge:
It just covers the basics but its a great refresher on both the basics of statistical variance which can occur as well as the mental attitude required in this game. There's a lot of similarities between trading and poker and if nothing else the video reminded me of what my attitude towards poker should be.
The graph below shows my results over the last few months and whilst the general trend is up, there are sustained periods of losses or break evenness and I think I need to handle these situations better. Truth be told considering the luck/variance that occurs in poker that chart is pretty good. In fact I feel personally I'm playing games to low and passing up potential profits in order to reduce the swings at the moment.
If I'm going to be managing large amounts of my own money, be it on an end of day or intraday trading system, I need to continue to develop the mental fortitude and recognize the 'process' rather than the mistake. I have a huge advantage over most because of my experiences but I'm not their yet.
I know I have positive expectancy in the games I play, just like in time I know I will have a positive expectancy in the market I trade. It's managing my emotions and continuing to play/trade well that is important. That and the 10,000 hours of experience.
I watched SPI Open today on the booktrader for the first time. It was pretty cool there was lots of activity for the first 15 mins or so and then it went pretty quiet after that. It's the first time I've actually watched something like this in action.
I assume the '1 @ 4853' means there has been 1 contract traded at the price of 4853. (Last Price)
My reading of the screenshot above is 'there are 8 contracts worth of bids who are happy to buy at 4853 but there are currently no sellers below 4854'.
Effectively if you want to buy you either go up and hit the market at 4854 or get in line at 4853. Similarly if you want to sell you can get in line at 4854 or hit the market at 4853.
I remember when I used to use IRESS at work we could see what the buyer depth was at each level for each buyer, not just a sum of the buyers on that level. Is it possible to somehow view that using Interactive Brokers as well?