My Hedged Fund - Another "Trend-Following" Post

[myhedgedfund]

Pictures below...

http://www.trade2win.com/boards/attachment.php?attachmentid=103294&stc=1&d=1298072000

 

[meanreversion]

Very promising start to the year, well done. At this moment, how much of your account would you say you have open to the market, i.e. if every single position suddenly went to its stop level, what % of your equity would you have lost?

 

[myhedgedfund]

Very promising start to the year, well done. At this moment, how much of your account would you say you have open to the market, i.e. if every single position suddenly went to its stop level, what % of your equity would you have lost?

Current "open to the market" risk as of today is 8.21%, meaning, the account will suffer an immediate loss of 8.21% if all technical stop-losses are hit at the same time. I track this number on a daily basis, and chart it on a weekly basis. It has been between 6.5 and 12% all year (see below).

 

[meanreversion]

Hmmm, you're running fairly 'hot' relative to my risk size ..... I think you mentioned at the very beginning that you would not expect to see a DD larger than 30%.... is this something you've calculated from backtesting? This is in no way having a go at you, I'm just trying to guage your risk parameters. For example, I bet 0.75% equity per market, up to a total of 10 markets, and would expect to see DD of between 20-30 at some point.

You have 20 positions open, but what risk as % of equity is each one?

Final question - do you know the annualised vol of your daily returns?

(Excellent thread by the way)

 

[myhedgedfund]

Correct Meanreversion; I expect a DD of no more than 35% as calculated from backtesting. It is actually a quite typical DD for trend-following systems, with the exception of the likes of Dunn, who have been known to experience a DD of up t0 60%, followed by expectacular returns of course (see sample performance chart from Dunn Capital Management).

View attachment Dunn Capital Management.pdf

Weighted risk as % of equity for my current positions listed below:

DBA 0.12%
ERX 1.81%
EWC 0.24%
EWJ 0.24%
FAS 0.34%
FCG 0.30%
IAI 0.13%
IYR 0.12%
MWJ 1.00%
PHO 0.16%
TNA 0.39%
TQQQ 0.95%
TYH 0.33%
UDOW 0.23%
UPRO 0.84%
XLB 0.14%
XLI 0.33%
XLY 0.12%
XME 0.23%
XOP 0.18%


I do calculate the annualized daily returns, but the data, while informative, has limited actionable value (for me based on my trading style). How do you use this data? I'm actually curious here...

BTW, thanks for the nice comment...

Boston

 

[meanreversion]

Annualised vol lets you calculate your Sharpe ratio fairly easily. You could even go one step further and compute your Sortino ratio (using the semi-standard deviation of negative returns).

Both these ratios are inherently flawed for various reasons (e.g. a short vol strategy will usually have a high Sharpe, but the actual risk taken is not properly reflected in the denominator.. Sortino suffers because you're arbitrarily throwing out positive returns, which implies no risk was taken to generate them).

None the less, Sharpe is the accepted ratio for the hedge fund industry, and in as much as you're a trend follower, I think there's validity in comparing your Sharpe with, say, Bluetrend or Winton, who both pursue similar strategies. The last I saw, Bluetrend had a Sharpe of 1.2 (they target 1-2) and Winton had a Sortino of 3 (which I believe to be the highest for the industry).

 

[meanreversion]

By the way, Dunn's drawdown is NOT typical of the larger systematic funds.. these days investors are wary about drawdown, anything north of 25% and they get nervous.

 

[JonathanPaul]

And dedicated Radianz lines to your brokers or were you going to rely on your bog standard ISP for uninterrupted net connection? Oh and did I ask if you wanted Bloomberg? That's $3000/month please.

going to get any change from your £1000/month?

</rant>

Where did you get that from???

Bloomberg doesnt cost anywhere near that/

 

[myhedgedfund]

Week 8: Was this a "Head Fake"?​

Most trends saw a significant deterioration this past week. The "fundamental" explanation revolves around the concerns associated with the problems in oil-rich Lybia. A significant amount of the oil production coming from the area was shut down, resulting in a spike in oil prices. This tends to concern most investors as any significant increase in energy costs is expected to drive up costs for most industries; this concern translates into lower stock prices (simplification at its best). The market did come back on Friday, so the question lingers... was this the beginning of "The Correction," or just the typical geopolitical event with its typical market effects? Time will tell.



So... Does trend following work with ETF's?

Many of you have read the great white paper titled "Does Trendfollowing Work on Stocks" published by Blackstar Funds. Their paper of course concluded that trend following does work with stocks, with a very telling chart (see below) showing how most trades result in "little losses" suffered early on in the investment, with a few longer-term "big winners" making up for the little losses and more.

Two months into the year, the Fund's chart is beginning to take on a similar look (see below). So does trend following work with ETF's? It's way too early... Time will tell.

Current Positions

I completed the following transactions this week:

- Reduced my positions in the Materials sector (XLB), Broker-Dealers Index (IAI), Midcaps (MWJ), Nasdaq (TQQQ), S&P 500 (UPRO), Industrials (XLI), Water (PHO) and Consumer Discretionary (XLY).
- I also shorted Chile (ECH) through a deep in the money option put.​

The spike at the end of the week resulted in several buy signals and I bought single blocks in the following:

- Bank Index (KBE), Grains (JJG), Copper (JJC), Media (PBS), Timber (CUT), Nuclear Energy (NLR) and Aerospace and Defense (ITA).​

Current positions listed in more details here.

Current performance

The "head fake" was hard on trends this week, especially on leveraged positions. The Fund declined by 4.89% during the week. Most holdings declined in value, with the exception of the Energy sector, which continues to shine. Chart with some relevant benchmarks below:

Current sector allocations below:

Questions and comments always welcomed and often answered...

Boston

 

[meanreversion]

Could you provide us with your vol of daily returns as well? Thanks

 

[Splitlink]

I believe that indices, whether they are of shares, commodities or whatever, indicate bubbly periods or bursting bubble periods. During these periods it is difficult to go wrong when going long or short.

Take a look at the run up to the financial crisis. One could have bought any of the Footsie shares and, in reverse, have sold any of them and made a profit.

The speculator puts his money into, practically, any rubbish in a boom and the experienced trader will sell that rubbish first, in a bust.

 

[meanreversion]

I believe that indices, whether they are of shares, commodities or whatever, indicate bubbly periods or bursting bubble periods. During these periods it is difficult to go wrong when going long or short.

Take a look at the run up to the financial crisis. One could have bought any of the Footsie shares and, in reverse, have sold any of them and made a profit.

The speculator puts his money into, practically, any rubbish in a boom and the experienced trader will sell that rubbish first, in a bust.

Interesting post, but not quite sure of the relevance to this thread (am I being dense??)

 

[Splitlink]

Interesting post, but not quite sure of the relevance to this thread (am I being dense??)

No, you are not being dense. I was reacting to an idea that crossed my mind at the time which was that, in general, all shares go up and down together in a general index trend movement and these trends work in cycles.

However, I think I am a bit out of my depth with this thread, so leave you to it, sorry.

 

[DashRiprock]

I believe that indices, whether they are of shares, commodities or whatever, indicate bubbly periods or bursting bubble periods. During these periods it is difficult to go wrong when going long or short.

Take a look at the run up to the financial crisis. One could have bought any of the Footsie shares and, in reverse, have sold any of them and made a profit.

The speculator puts his money into, practically, any rubbish in a boom and the experienced trader will sell that rubbish first, in a bust.

hi splitlink

one thing could be good if you can compare bubble-ness in stocks. like a beta for bubbles LOL

so then you can go normal beta zero but bubble beta plus or minus on wether you thinks its up trend or down cycle.

i read somewhere about trading stocks from searching the internet and on trading websites for retail punters all talking about the same stocks, that could be sortof the same thing.

 

[Splitlink]

hi splitlink

one thing could be good if you can compare bubble-ness in stocks. like a beta for bubbles LOL

so then you can go normal beta zero but bubble beta plus or minus on wether you thinks its up trend or down cycle.

i read somewhere about trading stocks from searching the internet and on trading websites for retail punters all talking about the same stocks, that could be sortof the same thing.

This is the trick that makes a trader successful but it depends on ,when buying, what makes a share cheap, but good value. I'm a bit of a fundamentalist on this subject. Shares that have little debt, all sorts of factors, and scattercharts have to reflect the components that you are studying. Scattercharts depend on the faith that you have in those components and we are all very different in our selection of those.

 

[meanreversion]

This is the trick that makes a trader successful but it depends on ,when buying, what makes a share cheap, but good value. I'm a bit of a fundamentalist on this subject. Shares that have little debt, all sorts of factors, and scattercharts have to reflect the components that you are studying. Scattercharts depend on the faith that you have in those components and we are all very different in our selection of those.

The issue I always had with fundamentals is that a) the data is usually old and already priced in, b) the data could be wrong (falsified etc), c) carrying out extensive research on a particular stock later causes behavourial biases to surface, i.e. staying long for too long despite price going down and d) difficult to cover several markets at the same time.

Trend followers claim it works because "human behaviour doesn't change and trends will always exist". We don't KNOW there will continue to be trends, nor do we know that the occasional trend will offset losses in all the non-trending markets. However, one apsect it does introduce into the trading is strict risk management, and it also helps get round biases, being mechanical.

But then again, the win rate is low, only rarely are there big up months, equity is volatile and so on. It's not for everyone, especially as the good "fundamental" traders will get in on the moves well before the system traders.

 

[Splitlink]

The issue I always had with fundamentals is that a) the data is usually old and already priced in, b) the data could be wrong (falsified etc), c) carrying out extensive research on a particular stock later causes behavourial biases to surface, i.e. staying long for too long despite price going down and d) difficult to cover several markets at the same time.

Trend followers claim it works because "human behaviour doesn't change and trends will always exist". We don't KNOW there will continue to be trends, nor do we know that the occasional trend will offset losses in all the non-trending markets. However, one apsect it does introduce into the trading is strict risk management, and it also helps get round biases, being mechanical.

But then again, the win rate is low, only rarely are there big up months, equity is volatile and so on. It's not for everyone, especially as the good "fundamental" traders will get in on the moves well before the system traders.

There has, always, been this argument, of course, but I feel that a company that has a policy of constant growth and low debt is a safer bet than TA selection. I always hark back to Next PLC because I have experience with it. Did you know that, in December 1990, Next PLC's share was 6.5 pence? I did not get in until much later, but I still got in at under 2 pounds. You won't find these shares in the FTSE 100 so TA stock selection should be confined to them. These companies have "arrived" (in fact, Next is one of them, now) so these are the value companies that make the index go up and down. Value companies have different components to growth companies and if you mix them in a scatterchart it is like mixing apples and oranges.


I could have better said "a company that has a policy of constant growth and low debt is a safer bet for TA selection than one selected for TA reasons, alone.

IOW, TA selection should be based on those companies with good fundamenbtals.

 

[myhedgedfund]

Week 9: A Quiet Week...​

It was a quiet week for "My Hedged Fund," both from an activity and final weekly performance, although intra-weekly volatility was higher than normal. The Fund had a volatility similar to that of the SPY which ended the week up 0.11%, with an 1.7% up day and 1.7% down day in between.

Current Positions

I completed the following transactions this week:

• - Sold the "Banks" position (KBE)
• - Bought Russia (RSX), Silver (AGQ) and Gold (DGP)
• - Added to my current Canada (EWC) position.

Current positions listed in more details here.

Current performance

The Fund was up 0.43% for the week, currently up 7.45% YTD. Energy and commodity sectors continue to trend well, helping create trends in those countries with significant energy and commodity industries such as Canada and Russia. An example of a chart with a nice trend (Energy) can be seen below:

Current chart with relevant benchmarks below:

Current sector allocations below:

Questions and comments always welcomed and often answered...

Boston

 

[michaelguan326]

Wonder if you considered adding another shorter term mean reversion system into your arsenal. Although TF systems do work well in the long run there are times when choppy markets decreases returns. From reading the whole thread you are expecting a drawdown of 35% but from what I have read in different articles backtesting dd is much lower then live, same goes for returns.

I actually just started into systematic trading myself and still in the process of system development, my TF system has around the same dd and an MAR of 1 on the dot, an actual fund manager suggests adding some uncorrelated shorter terms systems as they are known to help decrease dd even more.

In terms of MM, I have to say I am out of ideas. In my system, I have limited total sector risk and asset risk but don't know what else I can add to have a positive I impact...from your experience what do you suggest I look into to adjust risk?

 

[meanreversion]

Take two systems, both with positive expectancy, and add them together. If the correlation is negative, then risk adjusted return for the portfolio will be superior to either of the two, provided

1. Correlation of returns doesn't change adversely (a big if) and
2. Markets behave in a vaguely Gaussian fashion

In other words, a trend system could/should be complemented by a mean reversion system, but the trick is to find a consistently profitable system of this nature.

(Another option would be to consistently sell low delta out-of-the-money option spreads on the S&P)