Well, okay. Don't we have an interesting state of affairs? Equity markets seem to be in a holding pattern in many parts of the world. Lots of people think that this is the end of the decline and we should all be gearing up for a leg higher. An Elliott Wave B or the beginning of a new impulse count to fresh highs?
Well spring has sprung in Europe these last few weeks. And thank the Gods for that because the winter had become no fun, repeat NO FUN, at all. But I got through it, I prospered, I'm better stronger and faster for it and that is what counts. I just don't want to do it again too soon.
Here are some photos from a castle in Denmark, a short ferry ride across the Öresund from where we are. I'm ashamed to admit I don't remember the name!
And I was fortunate enough to find myself in Prague a couple of short weeks ago:
Enough gloating, eh? Hows about some market commentary?
The Fed cuts rates like crazy and gives Bear Sterns to JP Morgan Chase for a steal and we’re supposed to believe that the time is now suddenly ripe for equity markets to build a solid base and eventually head higher again? If the credit crisis really is as big as everyone (read: the mainstream and alternative media) is trumpeting, then it simply can’t be over. I mean, what are European banks doing in response to all of this?
Not that the pissy little Riksband is a bell-weather or anything, and I’m also open to the idea that they’re just plain stupid, but in February they raised rates another .25%. And just this week Danske Bank has reported that comments from the ECB are suggesting that they’ll go higher in response to “inflation concerns”. This is against the backdrop of media commentators using phrases like, remain elevated, when describing Euroland interest rate levels.
I don’t know. 4.25% don’t seem that high to me, but then again it’s all relative. The US has Missouri and Michigan and Alaska draining its welfare resources and growth prospects. The EU has France and Italy and elements of the German economy to contend with. It’s a different set of problems on a different scale. Maybe 4% is high, and 4.25% is very high…I dunno. What I do know is that locking rates at 4.5% for ten years still feels like the best thing I could have done.
And then we have Australia. There is simply no better place to have your cash right now :). Forget property. Leave it for those unfortunate people who need to buy for reasons other than financial gain/investment eg. moving in together, having kids, relocating for work etc. I know it’s a conventional trading wisdom to believe that prices are never too high to buy. In this case though, I’ll happily exchange wisdom for a good nights sleep.
I'm currently perusing opportunities in the Netherlands. I don't know anything about their market except that it's gone up a fair bit around Amsterdam in the last couple of years. I have a possibility to add a property to my portfolio and finance the mortgage payments at the equivalent of a 52% discount thanks to their income tax rebate arrangements. I’m contemplating how that could be used to greatest benefit.
I could save the difference into cash capital for the next opportunity, earning interest at Australian rates in the meantime. Or I could buy twice as much property as I could afford without the tax break. I could use the difference to pay down the mortgage faster. That would probably help the most if I were to refinance some years down the track and rent the place out.
The catch is that I have to own the property personally to receive the income tax benefit, but down the track I’d actually like to transfer it to a company structure in Sweden. And under Dutch law, unlike many other jurisdictions, I’m not sure that that translates to a capital gains event, but it does translate to a stampduty event, which I read somewhere is 6% of the sale price.
I also read somewhere that since 2005 gift tax has been abolished in Sweden for both individuals and companies. So strange as it may sound, in this instance, I would kind of hope for a flat market, wouldn’t I? Buy at a reasonable price, pay the transfer fees (6+%). Receive the tax break and hold the property for a few years, then gift the property into my Swedish company.
Of course at that point the Swedish company has to pay the transfer fees again. So the less the price goes up in the meantime the better...now isn't that an odd desire?
Still the point is, there are always positives to be found in any supposed negative situation. One just has to keep searching like a rat in a maze until you find the right (free) piece of cheese. As Mark Hudson is quoted as saying, “the economy is all about getting a free lunch, if it’s not free it’s not worth having because everyone else is getting it”.
So who in Australia didn’t take advantage of the first home buyers grant when it was at it’s highest? Or hasn’t sold a primary place of residence for a capital gain and avoided all tax liabilities? Just wondering :)
ASX.G
Sunday, April 27, 2008
What about some commentary on world markets?
Sunday, March 23, 2008
Only Fools Rush In
As the days become longer I feel inspired to write shit on the Internet once again. But someone forget to remind the weather that it was due to be spring soon. We got our first proper dump of snow here in Helsingborg on Monday.
A view from the Gorilla house above...
...and from within the Pålsjö forest...
...and of the Pålsjö castle, located about 20 minutes of casual walking from the Gorilla house.
"Wise men say only fools rush in, but I cant help falling in love with you Shall I stay would it be a sin if I can't help falling in love with you..." Elvis Presley
Elvis?!? Well this post isn't about love, sorry.
As people who read my blog will know by now...I'm no rusher-inner. I can spend a long time planning, and people can misinterpret this for indecision, fondly called analysis paralysis by many. Or perhaps some people misinterpret it for being dim-witted, or maybe just lazy. I don't know, and you guessed it, I don't care. I'm behind the wheel, not them.
There are two situations which I'm involved in at the moment where the idea that only fools rush in is prevalent in my mind. One is in the markets and relates to what has become one of the ultimate purposes of this blog ie. the release of a live trading system and a couple of live traded accounts (or maybe not, the jury is still out on whether I'll do this, but it's an idea I'm toying with). And the other context is in my day-job. I'm going to start there.
I was recently called to an emergency at a customer of mine. It required me to hop on a flight at very short notice and make myself present ASAP, to take care of things. I managed to get from location A to location B, including a short notice booking of an international flight, in a little over six hours. Not bad given the two hour check-in lead time and the hour drive/train ride to the airport and the 1 hour taxi ride out to site.
I'd been called, because I was a known entity who had only recently designed and successfully implemented a system (lets just call it that) for said customer. The result was a stable, dependable, and trouble free asset (as I like to refer to it), that this organisation could come to rely on to serve them for many years to come, if they so choose.
This emergency involved working on a legacy system that I had not designed, at another location, at very short notice, as this system had become grossly unstable. This directly effected the cash flow of this organisation, which measures it's annual turnover in billions of Euros. Yes, it always feels kind of special to be called on to help in such circumstances.
When I turn up I found what I would describe as nothing short of a disaster. I measure the success of my systems by the proportion of people that failure of my system will effect and the duration of that impact along with the criticality of that impact to the organisation. In otherwords, who couldn't do their work and for how long and what did it mean/cost? I know that five-nines, as it's refered to within the industry, translates to about 5 minutes and 15 seconds of downtime per year. The problem I was called in to address had had it's impact durations (yes, plural, many occurances) measured in days.
These people were deep in the proverbial. And yet when I turned up with my fresh eyes and possible insight, what do you suppose is the first thing they wanted me to do? Take a look around? Familiarise myself with their environment? Ask questions and get the facts? Try to actually troubleshoot and diagnose the problem? No, of course not. They wanted me to set about making changes to an already unstable environment. Thereby running the risk of potentially introducing new problems and not knowing whether those symptoms were part of my work or the original problem. Only fools rush in.
For those who might be interested in how a professional approaches troubleshooting, the methodology is always to try to cut the problem in half. The aim should be after cutting the problem in half three times to have localised it to the point where it can be rectified or worked around. Amatueurs who venture guesses at a solution may be right, from time to time. And here-in lies the risk of using a troubleshooting methodology.
If you go up against an amatuer, who would rather take a guess, and find that your 3 step approach is out done by their one step approach, you risk losing face and credibility. But over time, you'll win more than you lose. You only hope your decision to stick to your guns doesn't cause you to lose the big deals. This is what makes the game fun and can cause a few sleepless nights. Så är det (so it is).
Now to the markets. I have noticed, that when I backtest my system on out-of-sample data, using various different sample periods, that the start-up of said system is one of the riskiest times. Riskiest in terms of experiencing a large drawdown straight off the bat.
Many who have run LTTFs since the beginning of the bullrun have managed to weather the rapid corrections experienced on the ASX since June 06 with relatively small drawdowns. I say relatively because in my world anything greater than 25% is serious. 22% or less I reckon I can live with. That ground between 22% and 25% will probably hurt but I reckon I can live with it.
I've found with my own testing that if I activate my LTTF near a peak (or could it be, too far away from a significant trough), that it can experience some practically intollerable drawdowns. The immediate solution is to just reduce the leverage being employed, or change the stock universe to something less volatile like the ASX300. But have I mentioned, only fools rush in?
I think that these really large drawdowns occur because near the peak of a market, if you open new positions, rather than buying your equity curve, those positions have a high likelyhood of hitting the inital stoploss for a 1R loss. As Howard Bandy will tell you, this is the least desirable type of loss. Positions opened much earlier during the market trend will mostly likely be stopped out around peaks by either a trailing stop of sorts, or if you're really lucky a profit target. Both ways seems to be better for drawdowns.
Also by my reasoning such deep drawdown situations will be under represented by Monte Carlo testing. Unless my understanding of Monte Carlo testing is incorrect, I believe that it takes all potential trades from a sample period, and strings them together to create simulated, albeit artifical equity curves. Most people are very interested by these test results. And many seem to bandy it around like some kind of equity protection shield ie. "my system was Monte Carlo tested fifty-gazillion times and the max DD was x%".
The reason I believe that these large DD periods will be under-represented by many Monte Carlo testers is due to the fact that the relationship between trades is broken by randomised Monte Carlo testing. A random, artificial equity curve is worthless when the reality happens to be that stocks move up and down as parts of markets trends. This happens far more often than they move about randomly having no relationship to one another.
Curtis Faith explains this far better than I just did in his book, Way of the Turtle. He explains how his TraderBlox software keeps blocks of trades together to try not to under-represent the impact and likelihood of shock-events. Monte Carlo testing done incorrectly is not an equity protection shield for your trading system. Only fools rush in.
Start-date-bias, for both failure and success is a well understated determinant of many simplier LTTFs, IMO. A better approach would be to test using out-of-sample data, changing the start-date one day at a time, or if you are running a weekly system, using that interval. If I change my start date by a month sometimes the difference in maximum drawdown can change from 15% to 30%. If you can do this test, yet the Monte Carlo testing results don't demonstrate such a possibility then clearly the Monte Carlo testing lies.
Recently I've toyed with the idea of writing some code which prevents my system from buying an entire portfolio too rapidly. Have I already meantioned, only fools rush in. So that if an extreme peak does arrive soon after start up the index-filter I run the system with can kick in and prevent the opening of new positions which are only likely to result in a stack of 1R losses and a confidence crushing drawdown.
Another solution many people have for startup is to buy their equity curve. But I reckon that this could just equal a very expensive brokerage exercise as you open a full portfolio of positions pre-peak, then close them all out again post peak.
Disclaimer: a lot of this thinking is based on hunches. Whilst I subscribe to the belief that if I can test it, I don't need to rely on hunches, I also find I don't have all the time in the world, and eventually I need to find a solution that is good enough. As much as possible I want such a solution to be coded.
Remember, only f....alright, enough already.
ASX.G
"Wise men say only fools rush in, but I cant help falling in love with you Shall I stay would it be a sin if I can't help falling in love with you..." Elvis Presley
Elvis?!? Well this post isn't about love, sorry.
As people who read my blog will know by now...I'm no rusher-inner. I can spend a long time planning, and people can misinterpret this for indecision, fondly called analysis paralysis by many. Or perhaps some people misinterpret it for being dim-witted, or maybe just lazy. I don't know, and you guessed it, I don't care. I'm behind the wheel, not them.
There are two situations which I'm involved in at the moment where the idea that only fools rush in is prevalent in my mind. One is in the markets and relates to what has become one of the ultimate purposes of this blog ie. the release of a live trading system and a couple of live traded accounts (or maybe not, the jury is still out on whether I'll do this, but it's an idea I'm toying with). And the other context is in my day-job. I'm going to start there.
I was recently called to an emergency at a customer of mine. It required me to hop on a flight at very short notice and make myself present ASAP, to take care of things. I managed to get from location A to location B, including a short notice booking of an international flight, in a little over six hours. Not bad given the two hour check-in lead time and the hour drive/train ride to the airport and the 1 hour taxi ride out to site.
I'd been called, because I was a known entity who had only recently designed and successfully implemented a system (lets just call it that) for said customer. The result was a stable, dependable, and trouble free asset (as I like to refer to it), that this organisation could come to rely on to serve them for many years to come, if they so choose.
This emergency involved working on a legacy system that I had not designed, at another location, at very short notice, as this system had become grossly unstable. This directly effected the cash flow of this organisation, which measures it's annual turnover in billions of Euros. Yes, it always feels kind of special to be called on to help in such circumstances.
When I turn up I found what I would describe as nothing short of a disaster. I measure the success of my systems by the proportion of people that failure of my system will effect and the duration of that impact along with the criticality of that impact to the organisation. In otherwords, who couldn't do their work and for how long and what did it mean/cost? I know that five-nines, as it's refered to within the industry, translates to about 5 minutes and 15 seconds of downtime per year. The problem I was called in to address had had it's impact durations (yes, plural, many occurances) measured in days.
These people were deep in the proverbial. And yet when I turned up with my fresh eyes and possible insight, what do you suppose is the first thing they wanted me to do? Take a look around? Familiarise myself with their environment? Ask questions and get the facts? Try to actually troubleshoot and diagnose the problem? No, of course not. They wanted me to set about making changes to an already unstable environment. Thereby running the risk of potentially introducing new problems and not knowing whether those symptoms were part of my work or the original problem. Only fools rush in.
For those who might be interested in how a professional approaches troubleshooting, the methodology is always to try to cut the problem in half. The aim should be after cutting the problem in half three times to have localised it to the point where it can be rectified or worked around. Amatueurs who venture guesses at a solution may be right, from time to time. And here-in lies the risk of using a troubleshooting methodology.
If you go up against an amatuer, who would rather take a guess, and find that your 3 step approach is out done by their one step approach, you risk losing face and credibility. But over time, you'll win more than you lose. You only hope your decision to stick to your guns doesn't cause you to lose the big deals. This is what makes the game fun and can cause a few sleepless nights. Så är det (so it is).
Now to the markets. I have noticed, that when I backtest my system on out-of-sample data, using various different sample periods, that the start-up of said system is one of the riskiest times. Riskiest in terms of experiencing a large drawdown straight off the bat.
Many who have run LTTFs since the beginning of the bullrun have managed to weather the rapid corrections experienced on the ASX since June 06 with relatively small drawdowns. I say relatively because in my world anything greater than 25% is serious. 22% or less I reckon I can live with. That ground between 22% and 25% will probably hurt but I reckon I can live with it.
I've found with my own testing that if I activate my LTTF near a peak (or could it be, too far away from a significant trough), that it can experience some practically intollerable drawdowns. The immediate solution is to just reduce the leverage being employed, or change the stock universe to something less volatile like the ASX300. But have I mentioned, only fools rush in?
I think that these really large drawdowns occur because near the peak of a market, if you open new positions, rather than buying your equity curve, those positions have a high likelyhood of hitting the inital stoploss for a 1R loss. As Howard Bandy will tell you, this is the least desirable type of loss. Positions opened much earlier during the market trend will mostly likely be stopped out around peaks by either a trailing stop of sorts, or if you're really lucky a profit target. Both ways seems to be better for drawdowns.
Also by my reasoning such deep drawdown situations will be under represented by Monte Carlo testing. Unless my understanding of Monte Carlo testing is incorrect, I believe that it takes all potential trades from a sample period, and strings them together to create simulated, albeit artifical equity curves. Most people are very interested by these test results. And many seem to bandy it around like some kind of equity protection shield ie. "my system was Monte Carlo tested fifty-gazillion times and the max DD was x%".
The reason I believe that these large DD periods will be under-represented by many Monte Carlo testers is due to the fact that the relationship between trades is broken by randomised Monte Carlo testing. A random, artificial equity curve is worthless when the reality happens to be that stocks move up and down as parts of markets trends. This happens far more often than they move about randomly having no relationship to one another.
Curtis Faith explains this far better than I just did in his book, Way of the Turtle. He explains how his TraderBlox software keeps blocks of trades together to try not to under-represent the impact and likelihood of shock-events. Monte Carlo testing done incorrectly is not an equity protection shield for your trading system. Only fools rush in.
Start-date-bias, for both failure and success is a well understated determinant of many simplier LTTFs, IMO. A better approach would be to test using out-of-sample data, changing the start-date one day at a time, or if you are running a weekly system, using that interval. If I change my start date by a month sometimes the difference in maximum drawdown can change from 15% to 30%. If you can do this test, yet the Monte Carlo testing results don't demonstrate such a possibility then clearly the Monte Carlo testing lies.
Recently I've toyed with the idea of writing some code which prevents my system from buying an entire portfolio too rapidly. Have I already meantioned, only fools rush in. So that if an extreme peak does arrive soon after start up the index-filter I run the system with can kick in and prevent the opening of new positions which are only likely to result in a stack of 1R losses and a confidence crushing drawdown.
Another solution many people have for startup is to buy their equity curve. But I reckon that this could just equal a very expensive brokerage exercise as you open a full portfolio of positions pre-peak, then close them all out again post peak.
Disclaimer: a lot of this thinking is based on hunches. Whilst I subscribe to the belief that if I can test it, I don't need to rely on hunches, I also find I don't have all the time in the world, and eventually I need to find a solution that is good enough. As much as possible I want such a solution to be coded.
Remember, only f....alright, enough already.
ASX.G
Monday, March 17, 2008
Delft again next week...
Next week I’ll be travelling to Delft again. I’ll be staying here. You can see where I’m booked to stay and why I enjoy visiting this part of the world.
http://bb.go.nl/hotel/index.php?pageID=25&language=44
Today’s weather in Helsingborg is fine. The wind is blowing like hell and it’s snowing. Temperature, somewhere around the Big Orbison…that would be 0!
Saturday, March 15, 2008
Google Finance BETA - Cool feature
I was surfing around just now and stumbled across Google Finance Beta. I'm not ordinarily a news site guy, at least not a mainstream news site guy, but this cool feature where news articles were correlated to points on the price chart was worth noticing.
Friday, March 14, 2008
Firefox v3 (beta)
I was wondering when a new version of Firefox would arrive. V2 was starting to look dated alongside the latest IE and Vista. I just installed v3.04 (beta). I’m sure I’ll regret running a beta version, but nonetheless couldn’t help myself. It does look a let sexier, doesn’t it??
ASX.G
Posting from where ever!
This is a test post from my HTC. Sometimes I want to write some thing to the blog but by the time I get to my computer the idea or inspiration is gone. Apparently this facility can even accept photos up to 10meg.
Stay tuned for some 'live' posting!
ASX.G
Stay tuned for some 'live' posting!
ASX.G
Tuesday, March 11, 2008
Let's Talk About Trading
"You try to be greedy when others are fearful, and fearful when others are greedy."
”The stock market is a wonderfully efficient mechanism for transferring wealth from the impatient to the patient.”
Warren Buffet
”The stock market is a wonderfully efficient mechanism for transferring wealth from the impatient to the patient.”
Warren Buffet
Southern Sweden Update
The last couple of days in southern Sweden have been rather fine. So fine in fact that yesterday the temperature gauge in my car was showing 9.5 degrees and today I opened the window in my attic to get some fresh air while I worked. There are no leaves on any of the trees here yet, but one gets the impression spring is only around the corner.
I just finished a project with a large Swedish furniture company (no I can’t mention names, wink) which looks like it will lead to some further work down in Holland. Looking forward to that a lot. I also hear from my Dutch colleagues that as of the 1st of July smoking will be banned inside establishments in the Netherlands. As such, those wishing to experience purchasing and consuming contraband in a Dutch coffee shop must do so before this date.
The photos are of a farmhouse up for sale just outside Helsingborg. It was built in 1929 and is situated on 2 hectares (20,000 sqm.). Based on the photos, it’s a little bit too exposed for us. We’re looking for something with a bit more protection from the wind and some adjoining forest, or preferably even some forest on title. Elevation with a view would also be nice. And as I can get hay fever I’m not so keen on living beside a rape farm.
Those aforementioned things aside here is the kicker. The asking price is 1,750,000 SEK, or about $300,000 AUD at today’s exchange. I could be in the center of Helsingborg from here in about 15 minutes. Try that in Melbourne! And yes, Sweden also thinks that it’s had a property boom and that prices here are now expensive.
Lets Talk About Trading
Or lack of it. Well I claim that this is a trading and economics blog so at some stage I need to get things back on track. Short of making excuses for myself for not trading, I will say I’ve had the lion-share of my capital tucked snuggly away in a high-interest, at-call account back home in the land of respectable interest-rates. Ah, yes, by that I meant Australia.
So at around the beginning of October I finalized the framework for a medium to long-term mechanical trend-following system that I plan to take into the markets. For interest sake and nothing else lets take a look at what things might have been like if I’d activated this system (unleveraged) back then:
So at around the beginning of October I finalized the framework for a medium to long-term mechanical trend-following system that I plan to take into the markets. For interest sake and nothing else lets take a look at what things might have been like if I’d activated this system (unleveraged) back then:
Initial capital: $50,000
Ending capital: $46,301
Net profit: $7.4%
# of Trades: 23
Winners: 3 (13%)
Losers: 20 (87%)
Max. trade % drawdown: 20.0%
Max. system % drawdown: 12.3%
Ending capital: $46,301
Net profit: $7.4%
# of Trades: 23
Winners: 3 (13%)
Losers: 20 (87%)
Max. trade % drawdown: 20.0%
Max. system % drawdown: 12.3%
…versus…
Cash 7% compounded monthly: 2.3%
Conclusions: Aussie stocks have not been the place to be. Or my system sucks!
ASX.G
Tuesday, March 04, 2008
Vilket är dagens bästa mål?
Not that you can see it very well, but those little black stars under the white clouds in the top right corner of this web page mean that it's currently snowing in Helsingborg. That would be the second time in two nights and I am f%&king wrapped. I'm ecstatic about having to spend 15 minutes scraping snow and ice off my car until I can't feel my fingers. No honest, I am. I'm really, really thrilled about it. I mean why does one move to a Nordic country? If you answered: 'an abundance of slim, pretty, tanned, blonde girls' you'd be completely right. But if we have to endure such lengthy winters then at least make them mandatory white! There has been too precious little this year. Here's hoping for some car scraping tomorrow morning.
Dagens fråga. Vilken är dagens bästa mål? Dagen's bästa mål är lunch förstås!
You might see this poll on the hd.se web page I posted above. Translated. The day's question: What is the day's best meal? The day's best meal is lunch, of course! At least it is in Sweden. I have to confess I absolutely love the lunch culture here.
My boss asked me to the other week if I could mention some of the difference between how we did things in Australia versus how I see them done here in Sweden. He meant in a professional sense, but what I really wanted to share with him is how much I look forward to lunch each day. I've come to make it the centre-piece of my daily enjoyment of life. Let me explain.
First I need to introduce the concept of 'rikskuponger'. I'm actually not sure how to property translate 'riks'. Lets explain it by saying that the central bank in Sweden is called the riksbank. And properly spoken Swedish is called rikssvenka. Actually I just checked it up. They translate it as national.
Okay so you have national-kuponger (coupons). You buy them with your pre-tax crowns, and then pay fringe benefit tax on them. I think I worked out that I get about 1400 crowns a month for an outlay of say 1200 crowns. Not exact, but you get the idea. At today's exchange rate that's about $250 AUD of coupons $215 AUD of expenditure.
So where can you use these coupons? All over the place, but I believe that the key intention is that you use them to buy lunch, appropriately called, dagensrätt. Translated: the day's right. Practically all of Sweden is set up with these lunch restaurants serving different dishes of dagensrätt. The Swedish restaurants usually have serve food referred to husmanskost. Translated, everyday food. In Sweden this could be salmon, or sausage or a beef caserole or sometimes even pasta or a pie. Most places have at least two choices, some good places have a daily menu with 4 to 5 offerings. Like anything though, the quality varies.
After you've ordered you main meal you make your way to the salad buffe. Practically every lunch restaurant I've been to has a fresh salad buffe. In the depths of winter where you spend a lot of time inside being relatively inactive there is something satisfying about eating a big plate of salad for lunch everyday. This is included in the fare. Also included is a drink of something: a glass of light beer, or a soft drink or some sparkling mineral water along with coffee and something sweet for after.
The cost of all of this is typically about 65 crowns. If I allow for the tax benefit and convert to AUD that works out at about $9.90. Not bad. The portions are always a good size and you go home later that day feeling like you had a wholesome meal for lunch. So it's perfect when you are too tired to cook something for dinner because you don't feel like much anyway.
If I could take one thing from Sweden to Australia it wouldn't be the girls it would be the lunches.
ASX.G
Subscribe to:
Posts (Atom)
