mindmoneymarkets

Rotation model update

I've a lot of respect for Michael at Market Sci. Irrespective of the actual performance of the strategies on offer or blogged about, I have a lot of time for anyone prepared to put their record out there in public for all to see. It's relatively easy to research an idea, but showing its public forward performance is a different thing entirely. 

So in the interests of sharing successes and failures, I thought it was high time I updated on the performance of my rotation model and the lessons I have learned. 

The rotation model uses the excellent ETFRewind rotation tool. 

My chosen markets were:

UK Mid cap (FTSE 250). 
UK gilts. 
UK Index linked gilts

I used the Fidelity funds network platform which had free switching between funds. The model chosen was not the ideal model in theory and is limited by the choice of low cost funds available within ISA tax (UK tax exempt) wrappers. 

Effectively the rotation model is either half mid cap/ half bonds or full bonds. The kicker is the choice of bond to accompany the equity component in a bull market. In theory the model will keep steady in bear markets and not under perform too much in bull markets. 

Performance:

I started following the model at the point of the vertical line. 

The rotation model kept track with the benchmark until the Middle of August, after which the FTSE 250 took off with only a minor wobble with the resumption of the European sovereign crisis. 

In theory, the model as returned roughly 7%, while the benchmark has returned over 20%. 

So there is clear under performance since the summer. This is disappointing but not entirely unexpected, the aim was effectively mimic the performance of a 60/40 stock/ bonds portfolio with the greater stability of a 50/50 portfolio. In theory the model will rotate into the best choice bond for to accompany the equity component in a bull market to compensate for the mix being 50/50 instead of 60/40. 

Unfortunately in real life my performance has been below this, returning 2% since inception - Why?

In my last few weeks in Canada we hiked the west coast trail on Vancouver Island for 7 days, then came back to find my laptop fried along with my stored fidelity network login details. My physical files were already in shipping back to the UK. So in the process of relocating back to the UK and waiting for my shipped files to arrive/ login details to be reposted, it was around 4 weeks before I was above to get back in the saddle. Unfortunately this coincided with missing a big switch into the FTSE250. 

I've since discovered Dropbox which is a higher recommend backup and file sharing tool. 

I hadn't modelled what to do in this situation - I.e. the impact of a 4 week delay on the weekly rotation model. 

Should I have still switched into the FTSE250 and risk being late to the party as the trend reverses, or sit out and wait for the pullback?

Given the performance, I think you can guess which course I took!

In retrospect this was evidently not the best course of action. There are however a number of lessons I have learned. 

• A rotation model is about relative strength so waiting for pullbacks may not be the best option if a switch is missed. 
• Fidelity's on-line platform switches you out of a fund almost instantly, but can take up to a week to switch you back in again.
• Unfortunately the choice of funds and ETFs is limited in the UK compared to the US. Costs and transaction fees are unfortunately also higher, making some switching strategies inaccessible or too costly. 
• There are potential ways around this though such as changing the time horizon or using different vehicles for different assets. (E.g. Funds within an ISA for equities/ bonds, but use spread betting for gold). 
• I intend to continue with the rotation model, but will pull out some funds to put more into a PE/10 value based strategy. 

I hope post if of use. I shall update in another 6 months. 

Momentum Investing

There's so much great stuff out there on momentum/ switching methods that I thought I'd keep a blog page for myself and others noting the best articles/ resources. I'll update this page as I find more materials (As well as making it a bit more organised as more materials are added).

My Rotation Portfolio:

Using ETF Rewind's Rotation tool I set up a portfolio to experiment with a rotation model in May.

Original post here

Update 8th July 2010.

Since inception in May, the portfolio has returned a whopping 1%.
The benchmark FTSE 250 has returned 1.6% over the same period but with much greater volatility. 
The portfolio has been invested 50% in index linked gilts and 50% UK gilts with no changes so far. 

So far the model appears to be doing it's job with the index's outperformance only really coming today after the big jump in equity prices.

Given this week's rally, I wouldn't be surprised to see some changes coming soon with the model.

Rotation articles

• Relative Strength Investing - Meban Faber
• Simple ETF Momentum Strategy - CXO Investing.
• Simple ETF Momentum Strategy robustness/ sensitivity tests - CXO Investing.
  
• Cross asset style momentum - It's not just a stock market phenomenon. 
• MarketSci's series on the Fundztrader sector rotation model.
• AQR Capital research on momentum & value.
• Beta Surfing - CSS Analytics. Part1, Part 2 & Part 3. 
• Relative Strength 101 (CSS)
• Market Rewind's Live rotation models.
• WoodShedder on the ETF rotation model.
• Copstrat Monthly Momentum strategy. 
• Scott's Investments: ETF Relative Strength.
• Great summary of the theory behind momentum investing.

Rotation tools & resources

• ETF Rewind.
• http://www.etfreplay.com/ - Not tried this yet. Looks pretty funky.
• Moose Signals.
• Livermore Active Issues Index (CSS)
• Build an automated Sector Fund System in excel (Not tried this software yet, but looks promising)
• MY Best Funds - Tracking & Ranking of various mutual funds & ETFs.

The opposite end of the scale - Mean Reversion

• Mean Reversion in under performing markets - Mebane Faber
• Combining momentum with mean reversion in over extended trends.

If you find an interesting article in this field, do let me know and I'll add it to the list.

My Asset Rotation Model

Recently I've blogged about various asset allocation models with various academic and industry research backing up the approach as a sustainable market beater.

After numerous hours weighing up the true costs of various vehicles and analysing the best theoretical approach, I'm ready to put my money where my mouth is.

I've decided that Fidelity's stocks and shares ISA is the best option using their funds network. I then set about whittling down to the funds that have no load fee (initial charge), no switching fee and where possible, the lowest annual charge. 

Unfortunately this narrowed down the choice of assets available so commodities were unfortunately not something I could add to the portfolio. However an inflation linked gilts fund should in some way track the inflationary tendencies of commodities.

I used ETF Rewind's tool to check the correlation of the remaining funds.

 

These aren't the funds themselves but the tradeable ETF versions which the funds track closely (enough). The ETF Rewind tool couldn't hook up with yahoo data to get the actual fund performances, so this is a bit of a workaround.

Fund	Yahoo ticker for ETF
Fidelity MoneyBuilder UK Index	ISF.L
HSBC FTSE 250 Index Retail Acc	MIDD.L
Fidelity Funds - Euro STOXX 50TM A Inc GBP	EUE.L
HSBC American Index Retail Acc	IUSA.L
HSBC Japan Index Retail Acc	IJPN.L
M&G Gilt & Fixed Interest Income A Acc	IGLT.l
Fidelity MoneyBuilder Income (Gross)	SLXX.l
Legal & General Index Linked Gilt Index Trust Inc	INXG.l

I narrowed these funds down to those with the lowest correlation then had a good think about my priorities for this allocation of funds.

I decided I wanted to keep it simple and reduce risk by always having a hedge via the bond market.

So my final shortlist is:

UK Mid caps
Inflation linked Gilts
Sterling Gilts

During a bull market rally, the portfolio will be invested in mid caps and bonds and during an equity bear, it will switch to full bonds.

Here's the theoretical returns over the last few years using ETF Rewind's excellent tool.

 

This is a relatively conservative rotation strategy partly enforced by the lack of choice within Fidelity. It will no doubt under perform during bull markets, but I've already a fair amount of equity exposure from my pension. In the future I plan to allocate another chunk of cash to a Self Select ISA which will open up more possibilities and more equity based diversification.

I'll update with progress.

Regime Switching

Just had a read of an article from Rajiv Sethi who had this choice quote to make on regime switching:

"One ought not to expect markets to be efficient or inefficient, but rather to experience periods of relative efficiency that are interrupted from time to time by severe disruptions."

"Relative efficiency" I like that.

There is a real groundswell of academic/ industry and blogosphere research into the robustness of rotation/ switching strategies.

Only last night, I spotted another great piece of research from MarketSci, testing a Fidelity select rotation strategy. Adding a trend following filter as also suggested by Mebane Faber smooths and increases returns compared to the original idea.

There's a couple of missing pieces for me.

1. As already mentioned, trading costs have to be considered. The Fidelity select strategy makes use of the no switching fee facilities available. The good news is that this strategy is potentially workable from the UK thanks to the Fidelity Funds network. What I need to do next is research the number of funds available with no 'load' or switching fees to see if such a strategy would be workable within a tax wrapper. Another consideration is that although there may be no transaction costs, the funds still charge an annual fee and as I've blogged about here, this can materially eat into your returns. So your 'alpha' will have to be significant to overcome these annual fees. CXO address this issue from a US perspective here. 

2. Secondly, there has to come a point when, as Sethi points out, the momentum in anomalies/ strategies/ markets becomes a bubble. If you're a momentum switching model, there is always the danger of catching something at the top of the market. So I found it interesting to read this paper from Trimtabs which indicates that cash inflows into ETF's can be a great contrarian indicator. I wonder if adding this as a filter could act as a good bubble avoidance strategy. I'd like to test this,but at the moment am unsure how to get the data without it costing the earth.

Asset rotation models

Hi again After a month of family visits and pesky volcanoes, it's good to be easing myself back into some heavy trading research.

Some of my research has been on hold because I've been waiting for a good slice of time to get stuck into DV indicator's excel plug in.

I've enjoyed playing around with many ideas inspired by David over the last few months and hope the software will enable an acceleration of my research beyond my clunky excel efforts. In particular I'm looking forward to researching more on indicator/ system rotation.

Rotation

On the subject of rotation, I've spent the last couple of months playing around with ETF Rewind's Sector rotation tool and the more I research the idea of momentum switching, the more it make sense. 

The excellent CXO has a series of posts referring to academic and industry research on the effectiveness and persistence of sector/ index/ anomaly rotation models. 

There are a number of theories as to why medium term rotation strategies work.

According to AQR Capital Management

"Investments do not exhibit momentum over just any time horizon. For instance, we cannot say that the stocks that have performed best over the last five years will do so over the next five years. Indeed, on a five-year horizon we find the opposite effect in the data. Stocks that outperform for a long period of time will generally become expensive, and expensive stocks tend to under-perform less expensive stocks. However, the evidence does show that assets that have performed well over the last 12 months tend to do better over the next 3- 12 months than assets that have performed poorly over that same period"

Mebane Faber has recently put out a fascinating piece of research on the profitability of an asset rotation model.

it's well worth a read, as are the series of posts from David Vardi on rotation models. Here, here and here. 

A highly important factor appears to be the selection of assets with low inerrant correlation. E.g. choosing to trade the FTSE, a property ETF gilts & cash rather than the FTSE, CAC & DAX. 

This post is particularly useful in identifying the source of Alpha in rotation models. 

Naturally, this switching has a cost in the form of higher brokerage fees compared to buy and hold. None of the materials I've read so far take this into account. Before I put some of my funds to work in this arena, this is the final hurdle any model needs to pass.

Which reminds me, I need to report back on my final costs analysis & conclusions....

More soon