- This topic has 31 replies, 13 voices, and was last updated 6 years, 8 months ago by
.
- Posts
While I like the idea of this strategy it seems impractical for me at least. Though I have not tried it yet since I just learned of it, I suspect some issues with volume. For example for June it calls for 50% EIS. The average daily volume is 44000. If I just go with a half position it still would have me buying 6% of the average volume. I suspect many of the other countries are even worse. Is this mitigated by buying over several days for buying and selling?
For June no TMF needed?
When will this be added to the custom portfolio builder?
Nelson,
Some of the ETFs do have low liquidity; most are fine. We have tested this without using the lower liquidity ETFs and the results are pretty close; we are intending to give guidance on swap outs in any month when a low liquidity ETF comes up. Being patient with limit orders, as you implied, is necessary for any ETF that has more much than a couple of pennies spread on the bid/ask.We will be adding this to the custom builder; will update everyone on timing.
The current signal is up to date in the strategy section.
Kind Regards,
ScottMaybe I am looking at something wrong, but the numbers from 2012 and earlier on the table of trades don’t match up with the bar chart and stats in your article. Which numbers are correct?
More specifically, your bar chart at the top right corner of the page uses cumulative return numbers for 2010 – 2012 instead of annual return numbers. Were the wrong numbers also used for other calculations? Thanks.
I am curious how you came up with a 68 (or 66?) day look back period? How do different look back periods affect the results? I would be interested in seeing the stats on different look back periods you tested just to get a feel for the robustness of this strategy. Also, how does the strategy do using only the largest volume funds, say the top 20 or so? Thanks.
Derrick,
We look at stable results over different time periods for the settings, seeking forecastable plateaus rather than peaks. Generally small performance differences with moderate changes in lookback; radically different values produce strong differences but still generally positive outcomes. Strategy holds up well with small performance decreases with high liquidity ETFs.
How could you implement this strategy without using the 3x leveraged ETF TMF? Our 401K doesn’t allow leveraged ETFs.
Nice work, thank you for sharing. Could you be more specific on the allocation algorithm? Is the percentual split between top 4 ETF based on sharpe, performance or something more sofisticated? Or even trying all the possible combinations like US-TLT adaptive algorithm? Is there big difference between i.e. simple fixed ranking? Thanks
Frank & Team
This strategy was rolled out a year ago with the promise that this will do well, when the US equity market stalls. Well the US equity market did stall and delivered only 1.5% over the 12 months since the strategy was rolled out.
The Chart of this strategy for the last 1 year looks very similar to SPY (the benchmark) with much higher volatility despite the fact that TMF hedge was used. I suspect the correlation to SPY is larger than 0.6 or so for the last 12 months.
The return for the past 12 months seems to be a loss of 1.5% at a time when the SPY has made a positive return of 2%.
What went wrong?
Are you planning any tweaks? Or is this strategy designed to do well only in a strong US equity bull market?
Hello Sunil,
Good question.
The world Top 4 strategy is meant as a growth, non-U.S. strategy and diversifier. It is far more correlated to foreign developed and emerging equity indexes than U.S. equity. Foreign markets can grow significantly in the right environment. Last year was not such. If you have invested ‘naked’ in foreign equity markets in the past year, you would have been crashed. Remember, oil did touch below $30 and this influences many commodity producing foreign countries like Canada, Russia, Australia, etc. WT4 offered exposure to foreign counties while protecting from extreme losses by hedging with TMV (Treasuries),
To answer your question: What went wrong. The U.S. dollar rallied causing foreign markets as well as commodities to under-perform. If you believe this will continue you should not invest in WT4 but choose something more U.S. based, like UIS or the Nasdaq 100. But if you believe that commodity producing countries and foreign currencies will eventually outperform, then WT4 may be a good choice and a partial portfolio allocation could help diversify out of the U.S. dollar.Hi Vangelis
Is there a list ranking the top ETFs each month? The low volume on some of these is just to low for me and I’d like to move down to the next country ETF with more volume.
Many thanks
Jimmy VYes there is and you could do that. The ranking cannot quantify certain factors that our algo does (cross correlations, variable lookback periods) but it is close enough. Keep in mind that low volume does not always mean there is no liquidity Here’s an example: https://am.jpmorgan.com/blob-gim/1383272223898/83456/1323416812894_Debunking-myths-about-ETF-liquidity.pdf
Regards,
VangelisHi James,
As a subscriber, the ranking is part of our monthly signal. You can find it under “Strategy Signals” –> “Private: The World Top 4 Strategy signs for ….”Jimmy V,
I find the list of one-month ETF leaders at:
http://www.barchart.com/etf/performance/1month.php
a handy reference, although that includes all ETFs not just country-specific.
FW
If the purpose of this strategy is growth with diversification then why not remove the US based etfs from the model (QQQ and SPY) ..I don’t want to be adding more exposure to US markets when other models are already invested.
Secondly the TMF allocation might be dominating the trajectory at times. For market corrections may want to add short Emerging markets (EUM), and short S&P (SH).
Lastly, the etf selection on this model might lend better results with a shorter timeframe…shifting more weight to the 3 month performance. Case in point: RSX is not a holding at this time–is this because of mean reversion? or too long a timeframe used by the strategy?
Followup from my above post:
Since the strategy now uses the BRS top holding as the hedge instead of TMF this is clearly an improvement …however it comes with an unexpected consequence -> lower diversification when combined with other strategies. For example: a portfolio consisting of:
BRS, 10%
World top 4, 20%
Global sector rotation, 20%
Max Yield, 25%
Nasdaq 100, 25%This custom portfolio has a current allocation of 32% for december in JNK …this is due to the multiple strategies using BRS. This is a problem. When bonds are hot they will dominate the strategies using BRS for protection.
I’m not certain how best to fix this but one idea is remove BRS all together from World top 4 but add short Emerging markets (EUM), and short S&P (SH) as strategy asset classes to help with corrections with a max allocation on the short etfs of say 50%.
Another idea for further diversification would be to add the hedge at the portfolio level, if you use QuantTrader. In other words, limit the hedging instruments in the individual sub-strategies, and then add either the BRS or an own hedging strategy at the meta-level. If you prefer the fixed-weight allocations from our online portfolio builder, you can now simulate this much easier with QuantTrader using the “consolidated allocation” feature introduced in version 500: https://dl.dropboxusercontent.com/u/43364046/Logical-Invest%20QuantTrader%20versions/QuantTrader500S.zip
Anyone knows what’s going on with HEWC? Interactive Brokers and some other sites do not display current quotes (last one is from 3/23)… Volume is 0.
Hi Reuptake-
HEWC only trades on average 5,500 shares per day. It appears that there have simply been no trades since the 23rd, as strange as this may seem. Pull up a 1-year chart on IB and you will see this has happened several times in the past year. Yellow bars indicate days with no trades.
Thanks! Don’t you think that using such ETF in our strategies may cause problems? Eg. QuantTrader cannot work properly since it needs quotes from last close. And what about selling it? Is there some market maker who will buy it from us?
I agree with reuptake-
LI strategy backtests do not account for market impact from users entering and exiting the market, and they should not as this would be impossible to model accurately. But clearly with an ETF like HEWC which is a major LI website published strategy, there are likely many subscribers invested. LI subscribers alone will certainly impact pricing in a security as thin as HEWC as we all exit at once.
Perhaps Frank, Alex or Vangelis might want to consider a filter, dropping ETF’s from strategy universes that trade below a certain average daily volume of about 50-100,00 shares or below an average currency threshold of about $1,000,000 per day.
Accounting for user impact is one thing, using latest price is another. Recently volatility is rather high, and I think that if QT would use data not from last close but from close 2-3 days before it could select “wrong” ETF for strategy. It’s Wednesday now, and last quote is from Friday…
I’d rather trade non-currency-hedged version…
Thanks for the good discussion! HEWC volume indeed is small and has further dropped the last days. But technically speaking, QT has a close price for all days, e.g. there are no “empty” days which might create data issues, see screenshot.
Also there are constantly bid/ask available with a 5 cent spread, so trading the ETF with a limit order (or IB Algo), so trading 1600 shares, which corresponds to a 100k investment in the WorldTop4 (100k * 0.4 HEWC / 25 per share) should not be a problem. Keep in mind this – as other index pegged ETF by rule cannot deviate substantially from their underlying index – else this would be arbitraged away inmediately – so LI investors are not “making the market” here.
If liquidity still is a concern, there is no real alternative as a hedged MSCI Canada ETF, obviously EWC as a non-hedged ETF has much higher volume – at the cost of FX exposure.
Alex, thanks for reply, I was bit worried about bids, now I’m less worried. Still I think we can speak about “lack of data” issue, which can be significant when indexes can change as much as 3% during one day.
Good afternoon everyone,
While not sitting directly in front of my allocation percentages, I opted to invest in EWC instead of HEWC based on spread and liquidity. I understand this is the nonhedged version, so I believe I increased my hedge somewhere else to compensate. (And yes, I understand my substitute hedge is not the same as having the hedged version of EWC)
I have not had a chance to look at the returns of both to see if I had made a good choice or not.
Thanks
Mark
- You must be logged in to reply to this topic.