Forum Replies Created
- Posts
Thanks. For Tiingo and Yahoo, when exactly are the closing prices available for both the stock and crypto tickers?
Thanks Frank, I will send across. Enjoy your holiday! Use BTC-USD instead of SPY -> Makes sense, will try. But when I need to use QT for my other stock based strategies, do I need to change it back to SPY / 252 Trading days? Or should I run 2 separate instance of QT – one for crypto and one for stocks ?
Another question I have is for a weekly rebalancing strategy, currently, QT’s default rebalance period is Mon to Fri, 5 days assuming no holidays in between. For the purpose of my crypto trading strategy, could I change to say Wed to Tue (7 days), and I rebalance on the say day on Tue or the next day on Wed?
P.S. The “Notify me of follow-up replies via email” doesn’t seem to work. I checked the flag when I submitted my last set of comments but I didn’t get any email notification.
Hi Frank, hope all is well. I was playing around with this the last few days and was trying to create a pure crypto strategy. I was able to get the prices from Tiingo. They have pretty much prices of all the crypto tickers. Then, I have a risky strategy that chooses the top 3 coins and for the hedge, I use short eth/usd (will try btc/usd, but my model now is using short eth/usd in the hedge and uses an allocation setting of min. 50 and max. 80%), similar to what you guys use for the inverse vol ETF. In laymen terms, I want this strategy to find the best X/ETH and profit from that over time.
Based on the backtest, provided the LB is short enough (< 10 days) and the rebalance period is frequent (Weekly in my test), the result is quite encouraging. But a few sticking points:
– Price history of a lot of the alt coins is short, so this model might just overfit the results. I used the past 1 year as the period for optimizing the parameters.
– Quant Trader currently doesn’t support looking at prices daily and skip the weekend. It would be great if for crypto tickers, it can capture the weekend prices in the backtest as the swings are material over the weekend in crypto.
– Hedge – Short ETH when ETH goes into a big rally mode beating all the alts would not be a viable short. Instead of 50% hedged, one could just try to do say 40% to increase return, but that’s also not a good hedge in the traditional sense. If not ETH, perhaps the hedge could choose other alt-coin pairs that have liquidity and a high beta? Also, short ETH might not be available in a lot of accounts, but for those who have access to a crypto futures, they can get into this position via short ETH/USD futures, but one has to watch the funding.
Happy to share the file if anyone is interested in sharing how one might improve this strategy.
For the crypto portion of the exposure, do you guys use a listed product or manage the crypto separately? It’s easier to manage it all under one umbrella but listed options like GBTC and COINXBE have their own challenges. GBTC is trading at deep discount to NAV and COINXBE is only listed in certain European stock exchanges and is denominated in EUR. Have you guys considered using “crypto stocks” like Microstrategy and Marathon Blockchain as proxy?
XAUUSD would have been a good option, but it’s not available to the following folks for those who use Interactive Brokers, myself included unfortunately…
IB Spot Gold and IB Spot Silver trading is available to customers who are not legal residents of the U.S., Canada, Australia, Hong Kong or Japan.
Hey Vangelis, thanks for replying. Merry x’mas and happy holidays to you and to the team! I’ll pkay around with QT.
On a side note, do you guys do managed accounts for non-US people? I know you do for US folks.
Feel free to message or email me directly. I can contact you directly regarding my case and the specific restrictions I have. Thanks!
Dear Vangelis,
Essentially, I am trying to recreate a different version of the Aggressive Strategy or one with similar return / risk characteristics / diversification by keeping some form of the Nasdaq strategy which is a big component of that strategy. Currently, I can only do that by completely excluding the Nasdaq100 when I construct a portfolio using Portfolio Builder, but my allocation to the 3xUIS would be too high and the portfolio is too volatile. Hence, I am wondering whether I can modify the current nasdaq100 strategy to use ETFs only. My limitation is I cannot buy single name stocks, but I can do ETFs. Have you done any analysis on how your QLD + TMF strategy does vs N100 strategy? How should I go about creating a QLD + TMF strategy myself? I am a current subscriber of all strategies.
Regards,
HowardI can’t invest in single name stocks but I can do ETFs. Given how well this strategy performs over time, it really is a shame I can’t use this strategy in my portfolio. I considered whether I should just substitute QQQ with the top 4-5 stocks selected by the NASDAQ100 algorithm, but when looking at the unhedged performance of the strategy vs QQQ, the strategy beats the index by a good margin. Do you have any other suggestion for my case?
In addition, do you have any other suggestion on how I could optimize a portfolio without this strategy? I played around with the portfolio optimization tool but it doesn’t allow me to optimize with a return in mind and I ended up only getting results with just one strategy. The result doesn’t look correct.
I am wondering whether you can shed some light on the recent performance and allocation of the strategy. Over the past few months, YCS has been outperforming GLD, but yet, the model is recommending allocation of 70-80% to GLD (the underperforming asset) as opposed to weighting YCS or other long USD ETF higher. Intuitively, it doesn’t seem the strategy is following the USD strength momentum by overweighting the long USD etf. Looking forward to hearing your thoughts.
The BRS strategy has toggled its allocation between CWB-JNK and CWB-PCY over the past couple of months. Can you elaborate further on the driver behind the change? I know it is ultimately driven by the algorithm output, but do you think it’s detrimental to make the change? I am cautious of trading costs due to selling JNK to but PCY, and now after a month, I have to pay for the same trading costs to reverse the change. Please advise.
Thanks Vangelis. Re your point on high correlation, that would certainly be the case if you look back at the data long enough given all cryptocurrencies really just ran up over the past year. However, over the past few months, I am starting to see differentiation between various crypto assets, which was also written by another author here. Now given the infancy of this asset class, this could very much just be short term noise, and the space may not be mature enough for such investment approach. At pointed out in the article above, the optimized portfolio actually underperforms an equal weight portfolio in his test period. However, if we also throw USDT in the potential portfolio mix, intuition tells me there should be a way to come up with a rotation strategy that toggles allocation between the top crypto assets and cash, optimized for sharpe ratio or maximum drawdown constraints for instance. Happy to carry on the dialogue further. I am an existing subscriber of all the signals. What kind of subscription package can i get for QT? Please feel free to email / PM me.
Can QuantTrader be used to come up with a rotation strategy that toggles between the largest and most liquid cryptocurrencies (Bitcoin, Ethereum, Litecoin etc) and cash? That would be very a very interesting strategy to look into. Would be great if we can point me in the right direction.
Thanks,
HowardWill you consider releasing a leveraged version of GMRS or WorldTop4 strategy? As a subscriber to all strategies, I found having 2 options (leveraged and un-leveraged) for the Univeral Investment Strategy very helpful. Tks, Howard
LI team – The market cap for ZIV and TMF is only circa $86mm and it seems like the share outstanding has been decreasing over time. Do you still think it is liquid enough? Have you started looking at alternative instruments for this strategy just in case? Tks
I am trying to construct the portfolio for 2017 using ~3 strategies. While the custom portfolio tool is very helpful and powerful as it shows us the model risk/return of various strategy composition, I am trying to get a better understanding on the p&l driver for each strategy, i.e., what market will make a strategy works and what environment will make it not work as well until market normalizes. I have read the white paper, and read some of the forum posts, but would love to hear your advice on how best to get a better understanding on this.
For instance, 2015 was a tough year for various strategies given the macro events that drove volatility, especially the August 2015 cny depreciation sell off. Why is that? Is that because such events make momentum gauging difficult? I remember Alex mentioned that in one of his responses. But would love to hear your thoughts further. 2017 could be a bumpy ride given rates vol, various geopolitical events coming up etc.
Good luck investing in 2017!
Thanks,
HowardHi, I am a new subscriber. Thanks for all the great work so far. I am wondering whether the team has looked at using VIX/VXV as a signal as well. I am doing more research on when to go OW on the short vol ETF and have read from various sources that VIX/VXV < 0.9 also signal a good time to initiate a short vol position. Would be great to hear your thought on this topic.
