Ed Mamula.com » Narratives

Automatic 7 Allocation for May 2012

Ed Mamula — Mon, 30 Apr 2012 19:55:08 +0000

Another month where most of the models have arrived at the same decisions but in different proportions. I did introduce a mechanism to choose between VTI and IWM, and all models but the Classic chose to switch into VTI for May and out of IWM. Other than that, it is notable that all models but Classic are also allocating some money to bonds (IEF). This is the first time this year that any models other than Timid (which always holds some bonds) have held bonds, meaning that the models are getting a bit defensive…but the 10 year Treasury is sitting near 1.9% at the moment, so unless we make new lows on the 10 year yield, there won’t be much to gain from this IEF position. Finally, I expect to change (or eliminate) the Composite measure soon. The fact that it currently takes a vote from the Timid portfolio skews it too defensively.

May 2012	Timid	Vanilla	Smooth	Classic	Plus	Full Tilt	Composite
IEF	44	25	17.5		32	35	26.9
IWM				50			0.0
VTI	38	50	50		54	100	46.5
EFA							0.0
VWO							0.0
DBC							0.0
VNQ	18	25	32.5	50	26	65	26.6
GLD							0.0
Cash							0.0
Total	100	100	100	100	112	200	100.0

Automatic 7 TAA Results for February and March 2012

Ed Mamula — Sun, 29 Apr 2012 21:08:03 +0000

Results for February, March, and Year to date without much commentary. It’s much too early to start drawing conclusions, given that most of these systems have only been trading live since the beginning of the year. Being up 12.1% through the first 3 months of the year is nice I suppose, but I had to use leverage to get there, while an unlevered VT returned 11.9%. The bane of these models over the past2 and a half months has been that they are holding the Russell 200o (IWM) rather than the S&P 500 (SPY), and they have no provision for switching to large caps. Of course, I just noticed that Decision Moose made just such a switch from small caps to large caps in early March. For the moment, I will not tinker with the A7 models; I will simply keep allocating the “US Equity” portion to the Russell 2000.

System	Inception date	Feb 2012	Mar 2012	YTD	12 Mo
Ed’s Actual Results	Jan 2012	2.9%	-1.6%	12.1%

Optimized Systems:	Inception date	Feb 2012	Mar 2012	YTD	12 Mo
A7 Timid	Dec 2010	0.8%	-0.8%	4.1%	4.4%
A7 Smooth	Jan 2012	1.9%	-0.1%	7.6%
A7 Plus	Jan 2012	1.9%	-0.2%	7.1%
A7 Full Tilt	Jan 2012	3.7%	-0.3%	15.2%
A7 Composite	Jan 2012	0.9%	-0.3%	5.8%

Internal Benchmarks:	Inception date	Feb 2012	Mar 2012	YTD	12 Mo
A7 Vanilla	Jan 2012	-1.6%	1.9%	2.3%
A7 Classic	Dec 2010	0.7%	-0.1%	7.4%	12.1%

External Benchmarks:	Inception date	Feb 2012	Mar 2012	YTD	12 Mo
50% SPY / 50% IEF		1.6%	0.8%	5.2%	11.2%
GTAA		1.7%	0.3%	3.4%	-6.5%
VT		4.8%	1.2%	11.9%	-0.9%

Automatic 7 Allocation for April 2012

Ed Mamula — Fri, 30 Mar 2012 19:20:18 +0000

On balance, the systems are moving out of emerging markets (VWO), and into foreign developed (EFA) and REITs (VNQ). There seems to be very little going on in terms of interesting variations among the flavors of the systems, except for the fact that the Vanilla benchmark system is increasing its allocation to VWO and abandoning VNQ, which is the opposite of what most of the other systems have done.

January was great, February was okay, and March was a mild disappointment. February, March, and year to date results coming soon.

April 2012	Timid	Vanilla	Smooth	Classic	Plus	Full Tilt	Composite
IEF	44				8		13.0
IWM	28	50	50	50	46	100	43.5
EFA	11		17.5		21	35	16.8
VWO		50					0.0
DBC							0.0
VNQ	17		32.5	50	25	65	26.8
GLD							0.0
Cash							0.0
Total	100	100	100	100	100	200	100

Automatic 7 TAA Results for January 2012

Ed Mamula — Sat, 25 Feb 2012 16:29:58 +0000

The overview page has been updated to give a better description of each of these systems and my goals with them.

January was great from an absolute returns perspective, and the more aggressively allocated a system was, the better it did. A7 Classic was the strongest performer among the unlevered systems with a 6.8% return. I am very pleased to see A7 Smooth outperform the VT (All world stock index ETF) in January, because lest we lose sight of it, the basic goal of all of these systems is to participate in upside and mitigate losses.

My actual return of 10.7% is nice because this is the first month that I decided to crank up leverage in my personal account, and it’s always nice to start with a levered win instead of having to dig out of a hole. That being said, I underperformed A7 Full Tilt by 0.8% percent, which feels like quite a wide margin. (I used IYR calls instead of VNQ and Russell 2000 futures instead of IWM)

System	Inception date	Jan 2012	YTD	1 year
Ed’s Actual Results	Jan 2012	10.7%	10.7%

Optimized Systems:
A7 Timid	Dec 2010	4.1%	4.1%	14.5%
A7 Smooth	Jan 2012	5.7%	5.7%
A7 Plus	Jan 2012	5.3%	5.3%
A7 Full Tilt	Jan 2012	11.5%	11.5%
A7 Composite	Jan 2012	5.2%	5.2%

Internal Benchmarks:		Jan 2012	YTD	1 year
A7 Vanilla	Jan 2012	2.0%	2.0%
A7 Classic	Dec 2010	6.8%	6.8%	17.5%

External Benchmarks:		Jan 2012	YTD	1 year
50% SPY / 50% IEF		2.7%	2.7%	10.4%
GTAA		1.4%	1.4%	-3.8%
VT		5.5%	5.5%	-6.3%

Automatic 7 Allocation for February 2012

Ed Mamula — Tue, 31 Jan 2012 19:58:49 +0000

The allocations of all flavors of the Automatic 7 portfolio for February 2012 are below. I continue to target the “Full Tilt” allocation in my taxable account, and I am following the “Smooth” allocation in my retirement accounts. On balance, the systems are shifting away from IEF (US Treasuries) and into VWO (Emerging market stocks). A7 Vanilla, which is a simple TAA benchmark system, is once again the notable exception, allocating 25% of assets to GLD (Gold), and remaining very defensive with 50% of assets in IEF. January was a good month for practically everything, so it’s no surprise that all of these systems will show positive results. Full results for January will follow in a separate post.

Feb-12	A7 Timid	A7 Vanilla	A7 Smooth	A7 Classic	A7 Plus	A7 Full Tilt
IEF	39	50			8
IWM	31		50	50	46	100
EFA
VWO	12		17.5		21	35
DBC
VNQ	18	25	32.5	50	25	65
GLD		25
Cash
Total	100	100	100	100	100	200

*Corrected earlier typo in A7 Vanilla system; previously I showed 50% in IWM; this should have been 50% in IEF (the same as last month); error corrected 7:50PM 1/31/2012

Automatic 7 TAA Results for December 2011

Ed Mamula — Sat, 31 Dec 2011 22:19:15 +0000

Final update for 2011 using the “old” performance comparison table. A7 was down 0.6% and Mom and Pop was down 0.05% in December. For the full year, A7 was up 13.1% and Mom and Pop was up 6.8%. Among the benchmark portfolios, the “50/50″ (50% SPY and 50% IEF) porfolio was up the most, at 9.1%, and I will note that while both A7 portfolios have been stuck in reverse or neutral since the end of August, the 50/50 portfolio is at a new equity peak at the end of the year.

Okay, so there’s a lot to be happy about and a lot to work on. 13.1% returns during year when SPY was flat, VT was down 7.5%, and GTAA was down 7.5% looks pretty good. That being said, I think I can do a better job of following the system in 2012, and hopefully getting a higher return. I can also leverage the R code over at Systematic Investor and do more testing of system variations more efficiently.

Growth of $1
Month	A7	M&P	Msci	50/50	GTAA	VT
12/2010	1.073	1.030	1.017	1.016	1.046	1.077
1/2011	1.103	1.043	1.006	1.028	1.049	1.092
2/2011	1.146	1.063	1.036	1.045	1.077	1.124
3/2011	1.163	1.073	1.020	1.044	1.072	1.123
4/2011	1.206	1.101	1.070	1.069	1.111	1.172
5/2011	1.183	1.094	1.055	1.076	1.082	1.145
6/2011	1.154	1.081	1.039	1.064	1.060	1.130
7/2011	1.172	1.098	1.069	1.071	1.059	1.106
8/2011	1.253	1.110	1.082	1.066	1.005	1.026
9/2011	1.242	1.106	1.021	1.041	0.978	0.916
10/2011	1.209	1.092	1.027	1.091	0.981	1.019
11/2011	1.221	1.100	1.025	1.092	0.963	1.006
12/2011	1.214	1.100	1.011	1.108	0.968	0.996
Current Drawdown	3.2%	0.9%	6.6%	0.0%	12.8%	15.0%

Returns	A7	M&P	Msci	50/50	GTAA	VT
Last Month	-0.6%	0.0%	-1.4%	1.5%	0.5%	-1.0%
Year to Date	13.1%	6.8%	-0.6%	9.1%	-7.5%	-7.5%

A7 (Automatic 7 – EdMamula.com aggressive)

M&P (Mom and Pop – EdMamula.com conservative)

Msci (MarketSci TAA)

50/50 – 50% SPY / 50% IEF

(GTAA) Cambria Global Tactical ETF

VT – Vanguard Total World Stock ETF

Notes:

Interactive Brokers Tier I USD Margin Interest rate applied to leveraged portfolios (A7)

Automatic 7 Allocation for January 2012

Ed Mamula — Fri, 30 Dec 2011 20:00:03 +0000

Four of the Automatic 7 portfolios will be in REITs (VNQ), US small cap stocks (IWM), and 7-10 year US Treasuries (IEF). The A7 Vanilla, which is a simple benchmark that I track to see if there really is a benefit to all of this fancy tinkering, stands out from the others in a much more defensive position, allocating nothing to IWM, and allocating 25% in cash. A7 Classic, which is the only version geared for maximum absolute returns, stands out by being the most aggressive of the unlevered portfolios, by allocating nothing to IEF and splitting IWM and VNQ 50/50. I may regret giving myself the extra work of publishing all of these variations, but I think that over time, it will be quite instructive to see which flavors perform as expected and which do not.

A7 Timid		A7 Vanilla		A7 Smooth
IEF	47	IEF	50	IEF	20
IWM	37	VNQ	25	IWM	50
VNQ	16	Cash	25	VNQ	30
Total	100	Total	100	Total	100
A7 Classic		A7 Plus		A7 Full Tilt
IWM	50	IEF	27	IEF	35
VNQ	50	IWM	48	IWM	100
Total	100	VNQ	25	VNQ	65
		Total	100	Total	200

For my live trading in January, I will be following the A7 Plus allocations in my retirement accounts, and the A7 Full Tilt allocation in my taxable account. My live results may differ materially from A7 Full Tilt because I am getting this exposure by trading Russell 2000 futures, and March 2012 call options on IYR (another REIT ETF) in addition to holding IEF.

Results for December will be published in early January, along with a re-tooled version of the allocations page, so that we can keep track of all of these monsters.

2011 Roundup / 2012 preview

Ed Mamula — Thu, 29 Dec 2011 21:15:42 +0000

It’s never easy.

Running the Automatic 7 systems this year has been a great success and a great challenge. As 2011 draws to a close, all I can say is, it’s never easy to choose an allocation, and it’s never easy to deliver strong returns. I don’t suppose I should expect it to be.

Gold

Gold is an interesting buggar. I got it into the mix in mid – 2011 and proceeded to do all sorts of batty things that were not indicated by the system. At the end of the year, in my personal account, I lost about 4% of my equity trading in gold despite its theoretically positive contribution to A7′s 2011 returns. For 2012, I resolve to trade gold only within the confines of the system.

MLPs

I suspect that not having Master Limited Partnerships (MLPs) in the rotation is a blind spot and I would like to add AMJ to the rotation, but I have not as of yet. I will probably need to see a longer track record on this ETN (gasp, an Exchange Traded Note) before adding it in, if at all.

Following the plan

Besides gold, I had a couple of notable deviations from the plan in 2011. For one thing, I really levered up on treasury futures in September / October, and I also found myself doing put spreads on IWM in the 4^th quarter. Um, yeah, there was no reason for that, and it hurt overall returns. I will track my actual returns in 2012, just as I used to do when I traded forex. This tends to keep me following the system, as I don’t want to have to confess my baseless deviations in public.

Over thinking / tinkering

It is very difficult to avoid the urge to over think and over tinker with these systems…yes indeed, doing nothing is the hardest thing to do! I noticed today that a very simple version of A7 (which will be called A7 Vanilla) returned almost 20% this year. Going forward, I will use this simple version as a benchmark for a couple of fancier optimized versions.

2012 A7 system variations:

For 2012, I will be publishing seven flavors of A7 and stop tracking external TAA systems. External benchmarking is time consuming and it doesn’t add that much value to this site. I will link to other people’s systems as appropriate in the “Influences” section.

So here is the list of systems that I will be maintaining; please note, that as of today, this is not a signal service, and I do not provide investment advice. In the name of having an accurate, replicable track record, I will make every attempt to publish the allocations of all A7 variations in real time; however, if time does not permit me to do so, then I will publish only the A7 Vanilla portfolio in real time with the others being posted shortly thereafter.

A7 Vanilla – benchmark – simple parameters – not optimized

A7 Classic – benchmark – optimized for maximum absolute returns; no leverage

A7 Smooth– optimized for absolute return and smoothness of ride; no leverage; Goal: Beat A7 on a risk adjusted basis; does not attempt to beat A7 Classic on an absolute basis

A7 Plus– optimized for absolute return and smoothness of ride; sometimes uses leverage (primarily levers treasuries via the futures market); Goal: Beat A7 and A7 Vanilla on absolute and risk adjusted basis, and beat A7 Smooth on an absolute basis. A tall order! This is what I have been trading since August 2011, and I do believe if I ever ran a hedge fund or other money management firm, this would be the aggressive version available to the public.

A7 Timid (aka Mom and Pop) – optimized for smoothness of ride; Goal: Beat 50/50 benchmark on absolute and risk adjusted basis…and to use a cliché, “achieve stock-like returns with bond-like volatility”. Again, if I ran a hedge fund, this would be the conservative version available to the public.

A7 Full Tilt – optimized for smoothness of ride; utilizes 2x leverage at all times – Goal: Maximum return in a margin account.

A7 actual – Ed’s actual results in taxable account (attempting to mimic A7 Full Tilt, at least until the drawdowns cause a nervous breakdown!); will show deviation from A7 Full Tilt because of commissions, slippage, use of futures and options instead of margined ETFs, and operator error. Goal: Beat A7 plus on an absolute basis and be volatile enough for my high risk tolerance.

50/50 – benchmark — SPY/IEF – not optimized

Note, I use the term “smoothness of ride” to denote a proprietary optimization function that bears resemblance to a Sortino ratio.

Allocations for January 2012 will be published tomorrow. Happy new year!

Automatic 7 TAA Results for November 2011

Ed Mamula — Tue, 27 Dec 2011 19:05:09 +0000

Automatic 7 was up 1% in the month of November, while Mom and Pop was up 0.8%.

In the first 12 months of live trading, A7 was up 22.1% and Mom and Pop was up 10%. December is likely to be pretty close to flat (despite the swoon in gold…damn Federal Reserve ), leaving A7 up double digits for calendar 2011.

For next year, I will be doing less external benchmarking and more tracking of various flavors of my A7 portfolios.

Growth of $1
Month	A7	M&P	Msci	50/50	GTAA	VT
12/2010	1.073	1.030	1.017	1.016	1.046	1.077
1/2011	1.103	1.043	1.006	1.028	1.049	1.092
2/2011	1.146	1.063	1.036	1.045	1.077	1.124
3/2011	1.163	1.073	1.020	1.044	1.072	1.123
4/2011	1.206	1.101	1.070	1.069	1.111	1.172
5/2011	1.183	1.094	1.055	1.076	1.082	1.145
6/2011	1.154	1.081	1.039	1.064	1.060	1.130
7/2011	1.172	1.098	1.069	1.071	1.059	1.106
8/2011	1.253	1.110	1.082	1.066	1.005	1.025
9/2011	1.242	1.106	1.021	1.041	0.978	0.916
10/2011	1.209	1.092	1.027	1.091	0.981	1.019
11/2011	1.221	1.100	1.025	1.092	0.963	1.006
Current Drawdown	2.6%	0.9%	5.2%	0.0%	13.3%	14.1%

Returns	A7	M&P	Msci	50/50	GTAA	VT
Last Month	1.0%	0.8%	-0.2%	0.1%	-1.8%	-1.2%
Year to Date	13.8%	6.8%	0.8%	7.5%	-7.9%	-6.5%

Systems / Benchmarks:

A7 (Automatic 7 – EdMamula.com aggressive)

M&P (Mom and Pop – EdMamula.com conservative)

Msci (MarketSci TAA)

50/50 – 50% SPY / 50% IEF

(GTAA) Cambria Global Tactical ETF

VT – Vanguard Total World Stock ETF

Notes:

Interactive Brokers Tier I USD Margin Interest rate applied to leveraged portfolios (A7)

MyETFHedgeFund.com portfolios have been removed; the aggressive model is no longer freely available.

Automatic 7 Allocation for December 2011

Ed Mamula — Wed, 30 Nov 2011 19:20:17 +0000

Never a dull day, eh? As of last night, the system was prepared to drop gold and get more defensive. Thanks to coordinated central bank intervention on this final day of November, the models have shifted into a more aggressive stance…and not only that, it is a photo finish yet again on whether or not to include gold. I decided to leave it in, since the alternative was to hold two asset classes (small cap US stocks — IWM; US Treasuries — IEF) instead of three.

So, for December 2011, the model portfolio allocations are:

Automatic 7 (aggressive, leveraged): 55% IEF / 18% GLD / 50% IWM
Mom and Pop (conservative) : 53% IEF / 12% GLD / 35% IWM

I will continue to trade the “Automatic 7″ variety in my personal account.

Now my account is so small that I don’t even deserve to be called a small fish. I am a plankton on the ocean. I do wonder though to what extent this coordinated action was timed such that it would persuade the bigger fish and momentum chasers to jump in and start talking about the fabled Santa Claus rally.