QTAA Update: Conclustions and FAQs

Below we have updated our 2006 white paper.  While you can download the full 70+ page paper here, I’ve also chopped it up into a series of more digestible posts for the blog.  

The purpose in this paper was to create a simple-to-follow method for managing risk in a single asset class and, by extension, a portfolio of assets.  A non-discretionary, trend-following model acts as a risk-reduction technique with no adverse impact on return.  Utilizing a monthly system since 1973, an investor would have been able to increase risk-adjusted returns by diversifying portfolio assets and employing a market-timing solution.  In addition, the investor would have also been able to sidestep many of the protracted bear markets in various asset classes.  Avoiding these massive losses would have resulted in equity-like returns with bond-like volatility and drawdown.  Investors looking to tailor their portfolio may consider using alternate cash strategies, more assets in the portfolio, and alternative weighting schemes to find a portfolio that is right for them.

FAQs

Below is a compilation of frequently asked questions we receive on a regular basis about the broad global tactical strategies presented in our book and white papers.  While we cannot speak specifically to how we manage money, we can speak broadly to the strategies in this paper.

We try to be as open and honest about the benefits as well as the drawbacks of every strategy and approach we research.

Of utmost importance is finding an asset management program and process that is right for you.

Where did you get your historical data?  Can you send it to me?

We used Global Financial Data and our agreement does not allow us to share the data.  However, there are many free sources of data available including this post we did on a list of free data sources.

How do you update this model?  What do you mean by “monthly price”?

The model, as published, is only updated once a month on the last day of the month.  Market action in the meantime is ignored.  The published model was only meant to be broadly representative of the performance one could expect from such a simple system.

Have you examined an all-in version where you invest 100% of the assets in whatever asset classes are on a buy signal?

Yes, but this eliminates the benefits of diversification and exposes the portfolio to large risks when only a few asset classes are on a buy signal.  In addition, it introduces unnecessary transaction costs.  Returns are higher but with an unnecessary increase in risk.  Investors seeking higher returns can use leverage or employ some of the rotation techniques mentioned in this paper.

Do you rebalance the asset classes monthly?

Yes.  Although we show in the book The Ivy Portfolio that it is important to rebalance sometime, the frequency is not that important.  For buy and hold allocations we recommend a yearly rebalance in tax-exempt accounts, and rebalancing based on cash flows in taxable accounts.  Percentage tolerance bands are another option for rebalancing decisions.

Have you tried various moving averages?

Yes.  There is broad parameter stability from 3 months on out to over 12 months. 

I like the strategy and want to implement it, should I wait until the next rebalance?

We usually invest immediately at the rebalance point.  While this can have a significant effect on short-term results, it should be a wash in the long term.  Investors worried about the short term can stagger their purchases over a number of months or quarters.

Where can I track the strategy?

You can track the strategy on a number of websites including StockCharts.com.  We also have a tracking feature on the blog.

What about using daily or weekly data?  Doesn’t only updating monthly expose an investor to dramatic price movements in the interim?

We have seen confirming data for various timeframes, some superior, and some inferior.  The question is valid – but also consider the opposite scenario.  What happens to a system that updates daily where a market goes down fast, then reverses and goes straight back up?  The investor would have been whipsawed and lost capital.  We expect the timeframes to have similar performance over the long term.

What is the best way for an individual to implement the leveraged model?

This is tricky.  If your broker has reasonable margin rates then leverage is justified.  Interactive Brokers is consistently fair here. For investors familiar with the product, futures are a good choice. One can also use an all-in cross-market rotation system detailed in the paper.

Why are you taking credit for using the 200-day moving average model?

We’re not, and we are very upfront that trendfollowing models have been around for over 100 years.

For the rotation system you’ve written about where you purchase the top performer over the past 1, 3, 6, 12 month periods, are you simply using the mean of the 1, 3, 6, 12 month performance to calculate the top performer?

Yes.

Is the 10-month SMA crossover optimized for all (five) asset classes, or is it possible that different timeframes could work better for different asset classes?

Different timeframes will certainly work better (in the past), but what is unknown is the parameter that will work best in the future.  However, there is broad parameter stability across many different moving average lengths.

Have you ever tried adding gold to your model (or any other asset class)?

Yes, we use over 50 asset classes at Cambria – the paper is meant to be instructive.

Why did you choose the 10-month SMA?

Just to be representative of the strategy, and it also corresponds closest to the 200 day moving average. We chose monthly since daily data does not go back that far for many of the asset classes.

What software did you use to perform the historical backtests?

Excel.

I am trying to replicate your results with X (Yahoo, Google, etc.) database and my results don’t match.  What gives?

The indexes disclosed in the paper and book are obtained from Global Financial Data.  We cannot fact check all of the data sources to see how they calculate their numbers but make certain that the numbers are total return including dividends and income.  For Yahoo Finance one needs to use the adjusted numbers – AND make sure to adjust them every month (or record the new returns for that month), a tedious process.

How is your ETF different then the model you presented in the white paper and book?

We cannot address questions about our funds here.  Feel free to email us or visit the fund information page here:

http://advisorshares.com/fund/gtaa

Why the expansion from five to more asset classes?  Have you tested this increased granularity?

Yes we have performed extensive research in house that demonstrates that increasing the number of asset classes (and sub asset classes and industries) that are not perfectly correlated improves risk-adjusted performance.  You can find a brief post here on our blog.

What is the long term expected volatility and drawdown targets?  Is 10% maximum drawdown for the GTAA Moderate strategy reasonable?

While the historical volatility (7%) and maximum drawdown (-10%) are good targets, by definition a portfolios largest drawdown is always in the future.  We foresee the possibility of a 20% drawdown as a possible scenario.

How will trend-following asset allocation perform in sideways markets?

In general, a market that oscillates can be a poor market for trendfollowers do to whipsaws.  However, one that examines Japan – a market that has had very poor performance for the past 20 years – would find that a trendfollowing approach would have still added significant value.

What do you anticipate the long-term correlation of this strategy to be, with the S&P 500?

This is a difficult question to answer since correlations are inherently unstable.  However, since most of the portfolio is in equity-like assets we expect the correlation to be quite positive.  Historically the correlation has been around 0.7 with buy and hold and 0.5 with the S&P 500.

In your opinion, is this strategy diversified enough for an investor to have a large % of his assets in it?

The strategy is designed as a core holding and an “all-weather” portfolio designed to perform in any market environment.  The fund’s underlying holdings represent over 20,000 securities across the globe.

I can see using the strategy as a core holding, however, because it can hypothetically have a significant equity exposure, what is the largest allocation you would give it in a conservative account? 25%?

That depends entirely on the investor and their risk and return objectives.  The fund is targeting equity like returns with reduced risk, and depending on the investor that is appropriate for a 0% to 100% allocation.

It doesn’t look like it benefits the portfolio much to time the bond portion, thoughts?

Trading lower volatility bonds doesn’t have much of a benefit, but timing higher volatility bonds (junk, emerging, corporate) tends to work well.

Really Bad Months

Last month saw most asset classes decline, with the exception of US equities.  I thought I would put the declines in perspective – the below is since 1972 for the following main asset classes.

Most equity like assets have had worst months of over -20%.  Bonds, 7-15%.  Can you fathom that?

An old post on what to do after really bad months here.

 

Screen Shot 2013-06-13 at 11.03.46 AM

QTAA Paper Update: Weightings 10/10

Below we have updated our 2006 white paper.  While you can download the full 70+ page paper here, I’ve also chopped it up into a series of more digestible posts for the blog.  

EXTENSION 3 – WEIGHTING STRATEGIES

No two investors are alike.  Some investors value wealth preservation with low volatility above all else, while others can handle a 50% loss in an attempt at generating higher gains.

Below we look at a few different allocations that we will call GTAA Conservative, Moderate, and Aggressive.


GTAA Conservative   

This allocation broadly follows the allocation of GTAA Moderate, but with more in bonds (40% vs. 20%).  Cash is invested in 10 Year US Government Bonds.

cons 

 


GTAA Moderate 

This allocation is the same as mentioned in the prior Extension.

 agg

 

 


GTAA Aggressive 

This portfolio begins with the asset classes listed in the GTAA Moderate allocation.  It then selects the top six out of the thirteen assets as ranked by an average of 1, 3, 6, and 12-month total returns (momentum).  This method was detailed in our white paper “Relative Strength Strategies for Investing.  The assets are only included if they are above their long-term moving average, otherwise that portion of the portfolio is moved to cash.  We also include the effects of only investing in the top three out of thirteen assets.

Another extension we covered is to apply leverage to generate excess returns. An investor would simply invest twice as much in each asset class, and the maximum portfolio exposure would be 200% if all of the asset classes were on buy signals simultaneously. 

Note:  Implementing the leveraged model at many retail brokerages is not ideal due to prohibitive borrowing costs.  Leveraged ETFs likewise are not ideal due to large tracking error relative to the benchmark index.  An investor must be careful when pursuing leveraged returns.

 

Figure 20: Buy & Hold vs. Various GTAA Allocations, 1973-2012

 bh

Economic Fundamentals Suggest Higher Yield

I really look forward to Minerd’s purple charts of the week in their Macro View.  Today’s:

“Historically, the real yield on 10-year Treasuries has closely tracked the University of Michigan Consumer Sentiment Index. The correlation broke down, however, in 4Q2011, as a result of the Federal Reserve’s asset purchase program. The yield on 10-year Treasuries would be roughly 150 basis points higher than it is today if the market was not being distorted by Ponzi (uneconomic) buying.”

Chart-of-the-Week-06122013

QTAA Paper Update: Alt Cash Strategies 9/10

Below we have updated our 2006 white paper.  While you can download the full 70+ page paper here, I’ve also chopped it up into a series of more digestible posts for the blog.  

EXTENSION 2 – ALTERNATIVE CASH MANAGEMENT STRATEGIES

 

On average the tactical portfolio is invested in 30% cash.  This is a drag on the portfolio, and many investors employ other means to increase the yield on the cash portion of the portfolio using any number of funds or concepts.  Below we look at a simple method of taking on more duration risk by investing the portfolio in 10 year government bonds instead of Treasury Bills. 

 

Figure 19: Buy and Hold and GTAA Portfolios, 1973-2012

 19

 

An investor would have realized an additional 1.37% per annum in returns for marginally more volatility and drawdown – but how much of this is simply due to the major bull market in bonds?  We decided to examine a period of sharply rising interest rates from 1973-1981, and found that the benefit of taking on additional duration risk actually helped!

 

Figure 19: Buy and Hold and GTAA Portfolios, 1973-2012

 

19b 

 

QTAA Paper Update: Extensions 8/10

Below we have updated our 2006 white paper.  While you can download the full 70+ page paper here, I’ve also chopped it up into a series of more digestible posts for the blog.  

EXTENSIONS

Other than simplicity, there is no reason to only focus on five asset classes.  (Technically, we believe there are only four real asset classes:  stocks, bonds, commodities, and currencies. Everything else (like REITs) is a combination of the prior four.)  

At the same time, expanding a portfolio with allocations less than 5% of the total does not do enough to move the needle on the entire portfolio’s risk and reward characteristics. (This ignores derivatives and holdings with highly asymmetric payoffs). 

We also have the challenge that many asset classes and indexes simply have not existed for a very long time.  For example, we do not include TIPs, junk or high yield bonds, emerging bonds, foreign REITs, fundamental indexes, managed futures, currencies, or other asset classes we might otherwise consider.  However, thirteen asset class subgroups will likely cover the majority of the world that we would like to allocate to.

Below we expand the original portfolio from:

 a

 

 

…to include the following:

 b

 

We then take a look at the historical returns compared to the simple strategy of five asset classes.  As you can see, it improves returns about 150 basis points, likely enough to warrant increasing the assets in the portfolio.

Figure 18: Buy and Hold and GTAA Portfolios, 1973-2012

d

Floaters

A lot of people are moving into cash “substitutes” to increase their yield, likely with unintended consequences.  Below is a nice piece from Vanguard on floating rate bonds:

 

flot

QTAA Paper Update: Vol Clusters 7/10

Below we have updated our 2006 white paper.  While you can download the full 70+ page paper here, I’ve also chopped it up into a series of more digestible posts for the blog.  

WHY IT WORKS – VOLATILITY CLUSTERING

One of the benefits of a quantitative system is that it protects the investor from innate behavioral biases.  A discussion of some of the more insidious biases can be found in the Appendix.  Of course, this information is not only valuable for figuring out our own biases – other people’s mistakes leave the door open for us to soak up some of that elusive alpha.  As far as excess returns are concerned, for someone to gain, someone else has to lose.  People consistently make the same mistakes that are hard-wired into their brains, and they do so over and over again.

Humans use a different part of their brain when they are losing money than when they are making money.  We put together a 17-page white paper to address the topic called “Where the Black Swans Hide and the Ten Best Days Myth”.

Figure 18 shows the annualized returns and volatility for the five markets we studied in this paper.  On average, the returns are 60% lower and the volatility 30% higher when the market is below its 10-month simple moving average.  Commodities are the one exception where volatility is not higher when below the moving average, which makes intuitive sense.  Commodities are often driven by supply shocks that can result in price spikes.

2008 is a prime example with volatility levels in stock markets around the globe exploding to record levels.  However, this volatility has occurred after the markets already began declining. 


Figure 18: Volatility Clustering Across Various Asset Classes

18

 

QTAA Update: In Practice 6/10

Below we have updated our 2006 white paper.  While you can download the full 70+ page paper here, I’ve also chopped it up into a series of more digestible posts for the blog.  

PRACTICAL CONSIDERATIONS AND TAXES

There are a few practical considerations an investor must analyze before implementing these models for real-world applicability – namely, management fees, taxes, commissions, and slippage.

Management fees should be identical for both the buy and hold and timing models, and will vary depending on the instrument used for investing.  0.10% to 0.70% is a fair estimate range for these fees using ETFs and no-load mutual funds (obviously the lower the better).  Many all-ETF portfolios can be formed for approximately 0.1% to 0.3%.

Commissions should be a minimal factor due to the low turnover of the models.  On average, the investor would be making three to four round-trip trades per year for the portfolio and less than one round-trip trade per asset class per year.  Likewise, slippage should be nearly negligible, as there are numerous mutual funds (end-of-day pricing means zero slippage) as well as liquid ETFs an investor can choose from.

Taxes, on the other hand, are a very real consideration.  Many institutional investors such as endowments and pension funds enjoy tax-exempt status.  The obvious solution for individuals is to trade the system in a tax-deferred account such as an IRA or 401(k).  Due to the various capital gains rates for different investors (as well as varying tax rates across time, as well as the impact of dividends) it is difficult to estimate the hit an investor would suffer from trading this system in a taxable account.  Most investors rebalance their holdings periodically and introduce some turnover into the portfolio even for a buy and hold allocation – and it is reasonable to assume a normal turnover of approximately 20%.  The system has a turnover of almost 70%. 

Gannon and Blum (2006) presented after-tax returns for individuals invested in the S&P 500 since 1961 in the highest tax bracket.  After-tax returns to investors with 20% turnover would have fallen to 6.72% from a pre-tax return of 10.62%.  They estimate that an increase in turnover from 20%-70% would have resulted in an additional haircut of less than 50 basis points to 6.27%. 

There is some good news for those who have to trade this model in a taxable account.  The system results in a high number of short-term capital losses, and a large percentage of long-term capital gains.  Figure 17 depicts the distribution for all the trades for the five asset classes since 1973.  This should help reduce an investor’s tax burden.

 

Figure 17: Length of Trades for Timing Model, 1973-2012

 17

 

 

 

 

QTAA Update: Global Tactical 5/10

Below we have updated our 2006 white paper.  While you can download the full 70+ page paper here, I’ve also chopped it up into a series of more digestible posts for the blog.  

STEP 3 – GLOBAL TACTICAL ASSET ALLOCATION

 

Given the ability of this very simplistic market-timing rule to add value to various asset classes, it is instructive to examine how the returns would look in the context of an investor’s portfolio.   Here we introduce a version of the timing model we refer to as “Global Tactical Asset Allocation” or “GTAA”.   GTAA consists of five global asset classes:  US stocks, foreign stocks, bonds, real estate and commodities.  The returns for a buy and hold allocation are referenced as “Buy & Hold” or “B&H” and are equally weighted across the five asset classes.  The timing model also uses equal weightings and treats each asset class independently – it is either long the asset class or in cash with its 20% allocation of the funds.  Figure 12 illustrates the percentage of months in which various numbers of asset classes were held.  It is evident that the system keeps the investor 60%-100% invested the vast majority of the time (approximately ~80% of the time the portfolio is at least 60% invested).  On average, the investor is 70% invested.

Figure 12: Percent of the Time Invested, 1973-2012

 12

 

 

Figures 13 and 13b below present the results for the buying and holding of the five asset classes equal-weighted versus the timing portfolio.  The buy and hold returns are quite respectable on a stand-alone basis and present evidence of the benefits of diversification. 

 

Figure 13: Buy & Hold vs. Timing Model, 1973-2012, log scale

 

 13

 

 


Figure 13b: Buy & Hold vs. Timing Model, 1973-2012, non-log scale

 

 13b

 

 

However, the additional advantages conferred by timing are striking.  Timing results in a reduction of volatility to single-digit levels, as well as a single-digit maximum drawdown.  Drawdown is reduced from 46% to less than 10%, and the investor would have only experienced one down year of less than -1% since inception in 1973.  Figure 19 details the yearly returns, and post-2005 is highlighted as the out-of-sample period. 

 


Figure 14: Yearly Returns for Buy & Hold vs. Timing Model, 1973-2012

 14

It is possible that Siegel (or others) have optimized the moving average by looking back over the period tested.  As a check against optimization, and to show that using the 10-month SMA is not a unique solution, Figure 15 presents the stability of using various moving averages lengths ranging from 3 to 12 months.  Calculation periods will perform differently in the future as cyclical and secular forces drive the return series, but all of the parameters below seem to work similarly for a long-term trend-following application. 

Figure 15: Parameter Stability of Various Moving Average Lengths, Timing Model 1973-2012

 15

 

While it is instructive to examine the model in various asset classes, the true test of a model is how it performs out of sample in real time.  Since the paper was originally published in 2006 with results up to 2005, returns after 2005 should be seen as out of sample.  Figure 16 illustrates the returns for B&H and timing portfolios. 

 

 

Figure 16: Summary Annualized Returns for B&H vs. Timing Model, 2006-2012

 

 16

 

The model performed exactly as one would expect it to from historical data.  Namely, even though it only outperformed in three out of seven years, it beat buy and hold by over two percentage points per year, with much less volatility and most importantly to many investors, lower drawdowns.

Page 21 of 152« First...10...1920212223...304050...Last »
 
 
Web Statistics