- 09/08/2016 at 11:18 am #35433
We have discussed this before but we could look at the possibility again. Yes, we do have the capability to roll our results 1 day forward. I will try to post something here in the next few days.09/08/2016 at 11:55 am #35440
I’d like to follow up on IVANF’s very valid suggestions. I agree that either your performance tables need to track results on an open-to-open basis, or strategy signals need to be published at least an hour prior to the 4pm NY close. Most brokers require MOC (market on close) orders to be entered by 3:40pm EST. In my opinion it is probably easier for most users to trade on the open, but then most users should reasonably expect your performance tables calculated on that basis.
To consider the open / close difference over time is a wash, or to suggest any advantage will be arbed away, is doing a disservice to your subscribers.
Rolling the reporting date forward is an imperfect solution ss I believe it would detract from system performance. I believe it would be better overall to roll reporting BACK one day, so the subscribers are establishing new positions on the close of the last day of the month. My research shows that (because of new month inflows) establishing positions on the close of the last day of the month generates slightly better preformance than trading on the open of a new month and significantly better than trading on the close of the first day.09/08/2016 at 6:26 pm #35445
Thanks for the replies, I didn’t really think of the rolling back of 1 day like Gordon suggested , maybe that’s an easier implementation , but I guess that also runs the risk that the rankings could get changed depending upon price action on that last day of month between signal day and month end.
Whatever is the outcome of this discussion , I would like to highlight that not all subscribers are USA based , so if you were to publish signals say 1 hr before market close that wouldn’t really work well for people based here in Australia. Plus one needs time to put together the orders and rebalances , which can take some time, especially when one has multiple broker accounts. I also use CFD’s , and the platform doesn’t support MOC , so for those trades I put in limit orders the night before the USA market opens
So please keep this mind for any future changes to the timing of signal releases
Ivan09/08/2016 at 10:11 pm #35446
Sorry I forgot to add a feature request
Would it be possible to add the ability for users to select “Top N” number of symbols and see the back test results ?
ie instead of being restricted to 4 symbols , users can select the number of stocks they wish to trade
I think for larger portfolios especially ( I don’t have that problem yet…) the concentration on only 4 stocks could be problematic from a risk perspective ?
Your thoughts ?
Ivan09/09/2016 at 4:21 pm #35458
We are discussing how to best adapt our service to accommodate some of these very reasonable requests. We have a variety of clients both professional and private investors located in different time zones. Some are active traders, others are longer term investors that need to review, customize and trade across different accounts in different continents. We will try to satisfy everyone’s needs while we grow. In the near future we are looking into:
a. Being able to issue intraday bare-bones signals, at the end of the month.
b. Keep our current ranking, performance tables and commentary that many subscribers demand, for the next day as is done today.
c. Partner with a U.S. registered Advisory firm that will offer our strategies as managed accounts for hands-free investing.
d. Lower the $$ entry point for QUANTrader so that an individual non-proffesional investor can do what he wishes with our strategies (trade any day, customize parameters, backtest picking N top stocks/etfs, etc)
Some patience please as we are a small team.
Your ideas are welcome as always.09/09/2016 at 4:54 pm #35464
you can chose to invest in more than the Top4 shares, this is one of the reasons why we publish the ranking tables. We choose Top4 as balance between diversification and rebalancing effort, normally you would also diversify with a different strategy not to be too much in equities only.09/10/2016 at 3:25 am #35481
Thanks guys !
Yes we realise you are a small team , but its also a nice problem to have right ? Many customers who like your product
Looking forward to your new feature releases
Ivan09/18/2016 at 11:07 pm #35587
Hi LI Team,
can you tell me where I can find the individual contribution for each symbol in a given strategy ?
eg if I want to see the returns that can be attributed to say TLT in any given strategy , can I find that somewhere ?
Ivan09/19/2016 at 4:17 pm #35599
we´re currently not publishing this as an indicator, but the raw data is in the allocation return tables (Below Strategy Charts):
For the Nasdaq it would look like this:
Hist Return Distribution
Top 10: 180.04%
Bottom 10: -40.61%
Total: 263.98%09/20/2016 at 1:38 am #35605
Thanks Alexander ,
is there any way I can run the strategy back tests and exclude the returns from a given symbol ? ( eg TMF or TLT)
I’m just wondering what things would look like if I followed the stock allocations but when it comes to things like TLT or TMF to actually leave that allocation vacant
My rational is that since I use CFD’s , which entails paying interest on the full position size, maybe it isnt that smart for me to effectively be paying interest on a bond position ( also given the discussion around bond pricing in the future). In theory as long as the bond is paying more that the interest expense then I’m ahead , but in some cases the difference may be quite a small positive net return and if interest rates were to increase , then my borrowing costs increase and the bond return would likely decrease so its a net drag on my overall portfolio return. If that’s the case then I’m probably better off not taking the allocation, but I would like to see what the key metrics look like under this scenario
Do you think my logic is reasonable ?
Ivan09/21/2016 at 10:41 pm #35628
currently there is no way of running backtests excluding certain symbols as mentioned above. You might be right about your point of paying interest for bond CFD’s, wonder how that math looks like in detail (vs. reduced cost of capital as you do not invest in full instrument but only difference), would you share?
We will release out QuantTrader software at a more accesible price in future, then you couild do your own backtest.
Please give us some time though, have some things on the burner currently.09/26/2016 at 3:54 pm #35684
Hi-thinking of subscribing and would like to see a sample signal so an idiot like me can see if i can understand it. however the sample signal on this page is for the universal investment strategy-is it possible to see a sample signal for the nasdaq strategy? thanks!09/27/2016 at 3:52 pm #35697
Thank you for your interest in our strategies. Here’s a sample (September 2016 signals for Nasdaq100):
Private: The Nasdaq 100 Strategy – Investment for September 2016
This is the monthly email of the Nasdaq 100 Strategy.
Investment for September 2016:
20% AMAT, 20% NVDA, 20% SYMC, 20% CHTR, 20% TMF
The 2016 year to date (date of the strategy email) performance of the strategy is: 14.36%
The ETF ranking for this month is:
The table is showing the ranking based on the closing price of the last day of the previous month. The values in this table are for “reference only”. The current allocation can be quite different of the top ranked stocks in this table because QuantTrader considers additional data like mean reversion or cross-correlations which can not be displayed in this table. We may also exclude companies which recently had big price jumps because of a takover bids, earnings or any other special events. ETFs can be added or removed at any time during the year.12/18/2016 at 4:10 pm #37159
Hi, given a raising interest rate environment in the future – is it possible to publish a backtest of the Nasdaq Strategy by excluding TMF. The reason for my questions is in the past 20 – 25 years we have been in a falling interest rate environment and now we might have seen the bottom? – Will this change the way we use TLT/TMF. Same question for UIX 3X strategy – I am concerned about how TMF has completely negated the gains in SPXL from middle of the year…
Granted TMF may bounce but the question is will it trend down over the long run in the future…
-Nikesh12/28/2016 at 1:44 pm #37339
Have you considered backtesting using the 1x short Nasdaq etf and/or straight TLT as additional possible hedges?…model to choose only 1. I think the 3x treasury could sometimes washout gains and/or dominate the trajectory. The short etf would likely juice up the returns in 2008 or market corrections more so than 3x tlt.
Same idea with other models add the short s&p 500 1x option.
You must be logged in to reply to this topic.