"Hot Potato" anyone??

Discuss your favourite picks, broker, and trading or investment style.
User avatar
StuBee
Gold Ring
Gold Ring
Posts: 1410
Joined: 21 Sep 2010 11:08
Location: SW Quebec

Re: "Hot Potato" anyone??

Post by StuBee » 08 Aug 2016 21:54

inquisitive_one wrote:I started on May 16th - $26,984.55
June 16 - $27,555.19
July 16 - $29,038.40
Today - $28,723.56
He has been investing only in TD CANADIAN INDEX - E (TDB900) for the whole time. I'm using YTD instead of the last year just because I kind of started in the middle of the year.
The ETFs will always be cheaper, but mutual funds are easier. There is a 2% fee if you sell or change a fund within 30 days, but that's the only restriction

So, it seems that my friend made 6.4% profit.
I hope you do not think that this 6.4% yield since May 16th 2016 means anything!!! Five years of observation would be a minimum to evaluate any strategy (if it is a strategy)!!!

If you are actually taking the above numbers seriously (which you shouldn't) then what has happened in the last month (a drop in value) should be considered worrisome (which it is not).
"The term is over: the holidays have begun. The dream is ended: this is the morning."-C.S.Lewis, The Last Battle

ockham
Gold Ring
Gold Ring
Posts: 1426
Joined: 04 Apr 2006 21:50
Location: The Prairies

Re: "Hot Potato" anyone??

Post by ockham » 30 Nov 2016 20:09

I made (almost certainly) my last ever RRSP contribution a few days ago. I've decided to give the Global Hot Potato portfolio a tryout with the contribution.

I'll use four ETFs: XBB, XEF, XIU, and XUS. On the first trading day of each month, I will aim to place all the money ($25K), subject to reasonable rounding, into the ETF with the highest return in the immediately preceding 12 month period.

For the 12 month period ended Nov 29, ishares reports the returns as:
XBB: +3.41%
XEF: -2.07%
XIU: +15.11%
XUS: +8.87%

I will, therefore, aim to buy ~$25K of XIU tomorrow, Dec 1.

This is a psychological challenge for me. The very idea of momentum investing is deeply contrary to my instincts.

User avatar
AltaRed
Diamond Ring
Diamond Ring
Posts: 18103
Joined: 05 Mar 2005 20:04
Location: Ogopogo Land

Re: "Hot Potato" anyone??

Post by AltaRed » 30 Nov 2016 21:10

Keep us posted Ockham. Indeed, perhaps start your own thread "Ockham's Hot Potato Experiment" and track each month and advise. This goes against my psyche completely but I would like to be shown this could work.....
Imagefiniki, the Canadian financial wiki The go-to place to bolster your financial freedom

ockham
Gold Ring
Gold Ring
Posts: 1426
Joined: 04 Apr 2006 21:50
Location: The Prairies

Re: "Hot Potato" anyone??

Post by ockham » 01 Dec 2016 10:36

Train has left the station. I this morning purchased 1120 units of XIU at $22.28, for a total cost including commission of $24,963.55. Will now do nothing for a month, and then recalibrate.

AR: this thread isn't being used for any other purpose. For the time being, I'll just provide a monthly report here.

User avatar
deaddog
Gold Ring
Gold Ring
Posts: 2552
Joined: 19 Jan 2008 19:59
Location: Central BC/Arizona

Re: "Hot Potato" anyone??

Post by deaddog » 01 Dec 2016 11:23

ockham wrote:Train has left the station. I this morning purchased 1120 units of XIU at $22.28, for a total cost including commission of $24,963.55. Will now do nothing for a month, and then recalibrate.
What are you using as a benchmark? I assume it would be a 4 ETF portfolio but wonder how often you would rebalance that portfolio?
Most of our so-called reasoning consists of finding arguments for going on believing as we already do.( J.H. Robinson)

ockham
Gold Ring
Gold Ring
Posts: 1426
Joined: 04 Apr 2006 21:50
Location: The Prairies

Re: "Hot Potato" anyone??

Post by ockham » 01 Dec 2016 15:26

deaddog wrote:
ockham wrote:Train has left the station. I this morning purchased 1120 units of XIU at $22.28, for a total cost including commission of $24,963.55. Will now do nothing for a month, and then recalibrate.
What are you using as a benchmark? I assume it would be a 4 ETF portfolio but wonder how often you would rebalance that portfolio?
As stated yesterday upthread, I'll use four ETFs. On the first trading day of each month, all money is to be placed in the ETF which has the best return in the immediately preceding 12 month period.

I haven't defined a benchmark. My initial goal is to stick with the plan long enough so as to make comparisons with other more conventional "potato" portfolios meaningful.

User avatar
brad911
Gold Ring
Gold Ring
Posts: 1123
Joined: 29 Jan 2007 16:45
Location: London ON
Contact:

Re: "Hot Potato" anyone??

Post by brad911 » 01 Dec 2016 16:16

So why use 12 month vs. 6 month?
If the objective is to really use momentum and you're potentially changing monthly wouldn't a 6 month performance be more oriented to true momentum?
Ex: Trump's election, Pipeline bounce or oil supply cut won't really be reflected much in next months 12 month performance, but obviously is where the momentum today is heading. Just a thought as I wasn't sure if someone had addressed that already.
Triage Investing Blog - A Source for Value & Dividend Investing and Business Fundamentals

ockham
Gold Ring
Gold Ring
Posts: 1426
Joined: 04 Apr 2006 21:50
Location: The Prairies

Re: "Hot Potato" anyone??

Post by ockham » 01 Dec 2016 18:30

brad911 wrote:So why use 12 month vs. 6 month?
If the objective is to really use momentum and you're potentially changing monthly wouldn't a 6 month performance be more oriented to true momentum?
Ex: Trump's election, Pipeline bounce or oil supply cut won't really be reflected much in next months 12 month performance, but obviously is where the momentum today is heading. Just a thought as I wasn't sure if someone had addressed that already.
Perfectly good questions. My rationale for 12 months was simply that NormR used 12 months in the article upthread in which he defined the "hot potato portfolio."

As it happens, using 6 months instead of 12 would have dictated the same purchase today of XIU. Since this is an experiment, and its underlying idea is to test the proposition "The Trend is Your Friend", I'm open to arguments as to whether it's the 6 month trend or the 12 month trend that I should be testing.

DenisD
Gold Ring
Gold Ring
Posts: 2445
Joined: 19 Feb 2005 01:24
Location: Calgary

Re: "Hot Potato" anyone??

Post by DenisD » 01 Dec 2016 18:53

Six-month trend will respond to changes in trend faster. But it will have more whipsaw's and higher trading costs. By back testing, we can determine which period would have done better in the past. Unfortunately, there's no way to determine which one will do better in the future. :wink:

DenisD
Gold Ring
Gold Ring
Posts: 2445
Joined: 19 Feb 2005 01:24
Location: Calgary

Re: "Hot Potato" anyone??

Post by DenisD » 01 Dec 2016 21:47

I thought I'd run a backtest on P123 with a similar algorithm to ockham's. Unfortunately, I'm on the cheapest plan for the next few days. I can only go back two years. And Canadian ETFs are not available. So I'm using BND,EWC,VEA,VV.

The screen is rebalanced every four weeks. Slippage is 0.2%. The top screen chooses the ETF with the best 52 week price gain. The bottom is 26 week. The green vertical lines indicate an ETF change. The red line is the return of the screen. They blue line is the return of the S&P 500. Not pretty!
Attachments
P123W52.JPG
P123W26.JPG

ockham
Gold Ring
Gold Ring
Posts: 1426
Joined: 04 Apr 2006 21:50
Location: The Prairies

Re: "Hot Potato" anyone??

Post by ockham » 02 Dec 2016 13:24

@DenisD: if I read your screens correctly, (i) both the 6 month trend strategy and the 12 month trend strategy underperformed the S&P 500 by ~20% for the 2 year back-test period, and (ii) the 6 month trend strategy required one more rebalancing trade (6 vs 5). Is this right??

As you say, not pretty!!

DenisD
Gold Ring
Gold Ring
Posts: 2445
Joined: 19 Feb 2005 01:24
Location: Calgary

Re: "Hot Potato" anyone??

Post by DenisD » 02 Dec 2016 15:41

You are correct ockham. I'll try to list the individual trades tonight. Would be interesting to see how it did in 2008 - 09.

DenisD
Gold Ring
Gold Ring
Posts: 2445
Joined: 19 Feb 2005 01:24
Location: Calgary

Re: "Hot Potato" anyone??

Post by DenisD » 02 Dec 2016 21:46

Here are modified versions of the P123 backtest report for the above two charts. In each case, the screen switched to BND twice after an equity drawdown and missed a significant part of the recovery.

52 Week

Code: Select all

 Seq  Tran Dt   End Dt  Pos  Screen%   Bench%  Excess%  Screen$   Bench$    Costs
  27 16/11/28 16/11/30         -0.04     0.21    -0.25    90.69   111.29     0.00
  26 16/10/31 16/11/28  EWC     2.35     3.87    -1.52    90.73   111.06     0.35
  25 16/10/03 16/10/31         -1.49    -1.34    -0.15    88.65   106.92     0.00
  24 16/09/06 16/10/03         -0.69    -0.82     0.13    89.98   108.37     0.00
  23 16/08/08 16/09/06   VV    -0.15     0.14    -0.29    90.61   109.27     0.36
  22 16/07/11 16/08/08         -0.81     2.44    -3.25    90.75   109.12     0.00
  21 16/06/13 16/07/11          1.69     2.36    -0.66    91.48   106.52     0.00
  20 16/05/16 16/06/13          0.70     2.15    -1.45    89.96   104.06     0.00
  19 16/04/18 16/05/16          0.36    -1.05     1.41    89.34   101.87     0.00
  18 16/03/21 16/04/18          1.09     1.50    -0.42    89.02   102.96     0.00
  17 16/02/22 16/03/21          0.51     5.80    -5.30    88.06   101.43     0.00
  16 16/01/25 16/02/22  BND     0.61     2.08    -1.47    87.62    95.87     0.35
  15 15/12/28 16/01/25         -7.53    -7.29    -0.23    87.09    93.91     0.00
  14 15/11/30 15/12/28         -1.82    -1.75    -0.07    94.18   101.30     0.00
  13 15/11/02 15/11/30   VV     0.31     0.69    -0.38    95.92   103.11     0.38
  12 15/10/05 15/11/02         -0.50     6.04    -6.54    95.62   102.40     0.00
  11 15/09/08 15/10/05  BND     0.41     0.79    -0.37    96.10    96.57     0.38
  10 15/08/10 15/09/08         -6.35    -6.37     0.02    95.71    95.82     0.00
   9 15/07/13 15/08/10          0.01     0.14    -0.13   102.20   102.34     0.00
   8 15/06/15 15/07/13          0.59     0.66    -0.07   102.19   102.20     0.00
   7 15/05/18 15/06/15         -1.64    -1.70     0.06   101.59   101.54     0.00
   6 15/04/20 15/05/18          1.38     1.52    -0.14   103.28   103.29     0.00
   5 15/03/23 15/04/20         -0.55    -0.65     0.09   101.88   101.74     0.00
   4 15/02/23 15/03/23          0.20     0.20     0.00   102.45   102.40     0.00
   3 15/01/26 15/02/23          3.14     3.04     0.10   102.25   102.20     0.00
   2 14/12/29 15/01/26         -1.55    -1.69     0.14    99.13    99.18     0.00
   1 14/12/04 14/12/29   VV     0.69     0.88    -0.19   100.69   100.88     0.20
                                                         100.00   100.00
26 Week

Code: Select all

 Seq  Tran Dt   End Dt  Pos  Screen%   Bench%  Excess%  Screen$   Bench$    Costs
  27 16/11/28 16/11/30         -0.04     0.21    -0.25    90.37   111.29     0.00
  26 16/10/31 16/11/28          2.76     3.87    -1.10    90.41   111.06     0.00
  25 16/10/03 16/10/31         -1.25    -1.34     0.09    87.98   106.92     0.00
  24 16/09/06 16/10/03         -1.53    -0.82    -0.72    89.09   108.37     0.00
  23 16/08/08 16/09/06          3.53     0.14     3.40    90.48   109.27     0.00
  22 16/07/11 16/08/08          2.27     2.44    -0.17    87.39   109.12     0.00
  21 16/06/13 16/07/11         -0.10     2.36    -2.46    85.45   106.52     0.00
  20 16/05/16 16/06/13  EWC     1.65     2.15    -0.50    85.54   104.06     0.34
  19 16/04/18 16/05/16         -0.94    -1.05     0.11    84.15   101.87     0.00
  18 16/03/21 16/04/18   VV     1.06     1.50    -0.44    84.95   102.96     0.34
  17 16/02/22 16/03/21          0.51     5.80    -5.30    84.06   101.43     0.00
  16 16/01/25 16/02/22  BND     0.61     2.08    -1.47    83.63    95.87     0.33
  15 15/12/28 16/01/25   VV    -7.89    -7.29    -0.60    83.12    93.91     0.36
  14 15/11/30 15/12/28         -0.28    -1.75     1.48    90.25   101.30     0.00
  13 15/11/02 15/11/30         -0.14     0.69    -0.82    90.50   103.11     0.00
  12 15/10/05 15/11/02         -0.50     6.04    -6.54    90.62   102.40     0.00
  11 15/09/08 15/10/05  BND     0.41     0.79    -0.37    91.08    96.57     0.36
  10 15/08/10 15/09/08         -8.59    -6.37    -2.22    90.70    95.82     0.00
   9 15/07/13 15/08/10          0.12     0.14    -0.01    99.23   102.34     0.00
   8 15/06/15 15/07/13          0.33     0.66    -0.33    99.11   102.20     0.00
   7 15/05/18 15/06/15  VEA    -4.36    -1.70    -2.66    98.78   101.54     0.41
   6 15/04/20 15/05/18          1.38     1.52    -0.14   103.28   103.29     0.00
   5 15/03/23 15/04/20         -0.55    -0.65     0.09   101.88   101.74     0.00
   4 15/02/23 15/03/23          0.20     0.20     0.00   102.45   102.40     0.00
   3 15/01/26 15/02/23          3.14     3.04     0.10   102.25   102.20     0.00
   2 14/12/29 15/01/26         -1.55    -1.69     0.14    99.13    99.18     0.00
   1 14/12/04 14/12/29   VV     0.69     0.88    -0.19   100.69   100.88     0.20
                                                         100.00   100.00

User avatar
NormR
Gold Ring
Gold Ring
Posts: 3999
Joined: 18 Feb 2005 11:19
Contact:

Re: "Hot Potato" anyone??

Post by NormR » 02 Dec 2016 23:25

A few notes.

1. EWC didn't make an impact until near the end. So, the results should be similar to a similar U.S. method http://www.optimalmomentum.com/gem_trackrecord.html

2. The S&P500 is a poor benchmark for the approach.

3. It certainly doesn't outperform each - or a blend of all - of its constituent indexes all of the time.
[url=http://www.ndir.com]www.ndir.com[/url]

ockham
Gold Ring
Gold Ring
Posts: 1426
Joined: 04 Apr 2006 21:50
Location: The Prairies

Re: "Hot Potato" anyone??

Post by ockham » 03 Dec 2016 00:17

NormR wrote:A few notes.

1. EWC didn't make an impact until near the end. So, the results should be similar to a similar U.S. method http://www.optimalmomentum.com/gem_trackrecord.html

2. The S&P500 is a poor benchmark for the approach.

3. It certainly doesn't outperform each - or a blend of all - of its constituent indexes all of the time.
Your notes are noted, and understood.

Any thoughts/comments on a 6 month trend strategy vs. a 12 month trend strategy??

DenisD
Gold Ring
Gold Ring
Posts: 2445
Joined: 19 Feb 2005 01:24
Location: Calgary

Re: "Hot Potato" anyone??

Post by DenisD » 03 Dec 2016 01:21

One more little test to illustrate the role of luck in the outcome. Remember that my ETF screen checks for a change every four weeks. I ran the 26 week trend screen four times starting in different weeks in December 2014. So each run will check for a change during different weeks for the next two years. Here are the percent returns for the four runs from December 31, 2014 to November 30, 2016: -10.1, -10.5, 0.5, -10.1. Yes, one of them made a small profit. Do you feel lucky? :)

ockham
Gold Ring
Gold Ring
Posts: 1426
Joined: 04 Apr 2006 21:50
Location: The Prairies

Re: "Hot Potato" anyone??

Post by ockham » 03 Dec 2016 10:13

DenisD wrote: Do you feel lucky? :)
You neglected adding, "punk"!! (Maybe I should have averaged into the trend, not lump summed into it, or maybe......)

More seriously, in the hot potato's defence and as you know, start and end date bias will affect test results for any strategy, not just the hot potato.

longinvest
Gold Ring
Gold Ring
Posts: 1374
Joined: 10 Sep 2012 17:26
Location: QC

Re: "Hot Potato" anyone??

Post by longinvest » 03 Dec 2016 10:17

ockham wrote:
DenisD wrote: Do you feel lucky? :)
You neglected adding, "punk"!! (Maybe I should have averaged into the trend, not lump summed into it, or maybe......)

More seriously, in the hot potato's defence and as you know, start and end date bias will affect test results for any strategy, not just the hot potato.
Ockham,

You must have nerves of steel. I wouldn't have the emotional fortitude to invest into the Hot Potato and stick with it; it's really too hot for me.

Oh, and maybe you should have leveraged into it, if you really believe in it. :wink:

Good luck!
Bogleheads investment philosophy | Simple index portfolios | Lifelong Portfolio: 25% each of (domestic/international)stocks/(nominal/inflation-indexed)bonds | VCN/VXC/VAB/ZRR

ockham
Gold Ring
Gold Ring
Posts: 1426
Joined: 04 Apr 2006 21:50
Location: The Prairies

Re: "Hot Potato" anyone??

Post by ockham » 03 Dec 2016 15:12

longinvest wrote: Ockham,

You must have nerves of steel. I wouldn't have the emotional fortitude to invest into the Hot Potato and stick with it; it's really too hot for me.

Oh, and maybe you should have leveraged into it, if you really believe in it. :wink:

Good luck!
NormR counsels upthread that the hot potato portfolio is only for "aggressive, seasoned investors". I'm seasoned, but not aggressive, in fact risk averse. That's why my stake in the hot potato is a relatively small one.

CCP ran a series of articles this fall on smart beta ETFs. Momentum appeared to be a factor that gave a persistent outperformance advantage. Ghariton has been making the same kind of point here on FWF. The global hot potato portfolio offered a well-defined approach to testing the idea.

longinvest
Gold Ring
Gold Ring
Posts: 1374
Joined: 10 Sep 2012 17:26
Location: QC

Re: "Hot Potato" anyone??

Post by longinvest » 03 Dec 2016 15:55

Factors, on a pure level, are a mathematical construct. To have a pure factor exposure, one must implement a long/short portfolio, betting long on some securities and shorting the others. A pure value factor portfolio, for example, must invest into value stocks and short growth stocks. A momentum factor portfolio should do the same, investing into momentum stocks and shorting anti-momentum stocks. Anything else will provide a diluted factor exposure.

People run regressions (analyzing the past) to figure out the recent factor exposure of a portfolio (to guess the its future exposure by inferring that the near future will match the recent past), and adjust their portfolio accordingly. This is all done in a very mechanical way so that people can claim that this is "passive" investing. Right!

My opinion: It's all about selling the hope of beating the market. "Active investing" is not attractive anymore. Active fund managers get paid from the fees they collect on asset under management (AUM). With people fleeing active funds, they needed to attract "passive investors" back. They found the perfect formula to do so with factor investing.

Good luck!
Bogleheads investment philosophy | Simple index portfolios | Lifelong Portfolio: 25% each of (domestic/international)stocks/(nominal/inflation-indexed)bonds | VCN/VXC/VAB/ZRR

OhGreatGuru
Gold Ring
Gold Ring
Posts: 1077
Joined: 27 Mar 2010 16:01

Re: "Hot Potato" anyone??

Post by OhGreatGuru » 04 Dec 2016 20:30

inquisitive_one wrote:A friend mentioned the "hot potato" theory to me, as an experiment on his RRSP account. It seems like a hybrid of passive & active investing. This is what the portfolio looks like:

25% TD Canadian Bond Index - e {TDB909} -> MER 0.50%
25% TD Canadian Index - e {TDB900} -> MER 0.33%
25% TD US Index - e {TDB902} -> MER 0.35%
25% TD International Index - e {TDB911} -> MER 0.51%
...
I am utterly confused (or you are) The 25/25/25/25 allocation is a global couch potato portfolio.

The "Hot Potato" theory you describe shouldn't be allowed to use that name, because it seems to me to be no more than a variation of market timing using index funds. (See the article in the link posted by DennisD /save/investing ... portfolio/

And market timing completely opposes the whole theory of couch potato investing.

User avatar
NormR
Gold Ring
Gold Ring
Posts: 3999
Joined: 18 Feb 2005 11:19
Contact:

Re: "Hot Potato" anyone??

Post by NormR » 04 Dec 2016 20:59

OhGreatGuru wrote:The "Hot Potato" theory you describe shouldn't be allowed to use that name, because it seems to me to be no more than a variation of market timing using index funds.
I think the name is pretty good, if I do say so myself. It expresses some of the essence of the strategy while being a bit cheeky and provocative.
[url=http://www.ndir.com]www.ndir.com[/url]

DenisD
Gold Ring
Gold Ring
Posts: 2445
Joined: 19 Feb 2005 01:24
Location: Calgary

Re: "Hot Potato" anyone??

Post by DenisD » 04 Dec 2016 23:08

ockham wrote:More seriously, in the hot potato's defence and as you know, start and end date bias will affect test results for any strategy, not just the hot potato.
Indeed. When I started running some long backtests a few years ago, I was surprised at how much varying the start date by even a week or two could change the long term results. Frequently 5 or 10%/year. It helped to remove my rose coloured glasses.

I've started to think that the phrase – past performance does not guarantee future results – should be changed: past performance does not guarantee past or future results. :wink:
NormR wrote:I think the name is pretty good, if I do say so myself.
:thumbsup:

ockham
Gold Ring
Gold Ring
Posts: 1426
Joined: 04 Apr 2006 21:50
Location: The Prairies

Re: "Hot Potato" anyone??

Post by ockham » 05 Dec 2016 09:55

DenisD wrote:
I've started to think that the phrase – past performance does not guarantee future results – should be changed: past performance does not guarantee past or future results. :wink:
NormR wrote:I think the name is pretty good, if I do say so myself.
:thumbsup:
My hot potato (I rather like the name) experiment is already worth the try. "Past performance does not guarantee past results" is an investment maxim to remember. Love it!

User avatar
NormR
Gold Ring
Gold Ring
Posts: 3999
Joined: 18 Feb 2005 11:19
Contact:

Re: "Hot Potato" anyone??

Post by NormR » 15 Dec 2016 14:59

A quick return update: A Hot Potato feast
[url=http://www.ndir.com]www.ndir.com[/url]

Post Reply