userque & NUGT/DUST

T+3

The system is trained on historical data. It is only trained for forecasting two days ahead.

T+3 Simply peeks ahead additional days (from the already forecasted two days) to see that price action. BUT, the system was not trained to forecast those additional days. The forecast could be different if the system were to be trained on the additional days.

T+3's performance is updated as well when the 'system' itself is updated.

T+3's accuracy is a function of the system's ability to forecast, as well as the (un)predictability of GDX. The two are difficult to differentiate, if possible at all. I.e. The longer the outlook, the less accurate the forecast. For all I know, GDX may be essentially, "unpredictable" past a certain number of days into the future. This is why I stick with two days. I could do one day, but I want to avoid trading against the trend, if possible. Two points (minimum) are required for a line: Two days (minimum) are required to define an eod trend.

Finally, as mentioned, if GDX is so volatile, why would one even want to rely upon a long term forecast when you don't have to.

10 Minutes before Closing

The system uses, inter alia, the current day's HIGH, and LOW prices. I could forecast sooner, but that increases the chances that the HIGH or LOW will change. The system "compares" historical eod highs and lows. It is trained on historical eod highs and lows. Giving the system highs and lows from noon will skew the results, as the system expects to receive eod data.

Giving it data from 15:50 is not eod, but it's better than giving it data from 15:40.

The system is robust and a 15:40 forecast, most times, will be the same as the 15:50 forecast, I suspect.

Let me know what time you would prefer to have the forecast and I'll see what I can do.

:smile:

_______________________________________________________________
DISCLAIMER: Userque's trading signals are not affiliated with TSP Talk or Buy Low Sell High, Inc. The information is for educational purposes only! The information is not advice or recommendations. The information may be revised at anytime. Userque does not give investment advice. Do not act on this data. Do not buy, sell or trade any stocks, ETF's, indices, funds, or any other instruments/vehicles based on this information. Userque may trade differently than discussed or posted in this forum.
 
:blink: LOL. I guess my system is not for everyone. :)

My system does give direction. ENTER NUGT means the system is long. ENTER DUST means the system is short.

Sideways is chop.

Again, it'll swing 2-4 or more days OR it'll day trade.

It trades at open or close, not intraday. Still a day trade.

It doesn't have predetermined levels at which to enter or exit. Still a day trade. That's just the way I designed it, based upon how I think the markets work. Others may design systems based upon their theories.

See prior (a day or two or so ago) posts regarding updating the system.

Some folks trade with levels; some don't.
Some day trade; some don't.
Some day and swing; some don't
Some update their systems; some don't.

To each, his or her own. Right?


:blink:

Just following along this thread...no, I'm not trading actual money, and yes, I can see both RMI's and Q's points ! Day trade...swing trade..."trending"...whatever !!! :smile:

IMHO...this is Q's "system"...he has been MORE than open about when he feels the need for a change or an update to his system, and he's shared that for all to see...as well as his running results, either great or ugly ! I don't understand the underlying parameters of his system, but my feeling is...it's a work in progress ! That's why Q is sharing it with us all, instead of charging us 100's of dollars for his "system" !!!

Again...it's YOUR decision whether or not you follow his "system", whether by day or not !!! Just use it at your own risk...and I don't mean to imply that in any negative way... !!!


Stoplight...
 
Q,

I agree with Stoplight. It is your system, you do whatever you think is right. Plus you have the advantage of knowing the ramifications of publishing your calls early vs late. With that said my thinking is most of the time the daily highs and lows are made throughout the day and not usually at the EOD. It would seem to me that statistically making a call 15 to 20 minutes before EOD would most likely have the same input data as at the EOD. However, I must admit I don't know how long it takes your system to "crunch" the data once it has been entered.
 
Thanks Stoplight,

I appreciate that. I think RMI is trying to help me in some type of way; maybe we're just not understanding what the other is saying right now. :D
_______________________

Proverbs 27:17 Iron sharpeneth iron; so a man sharpeneth the countenance of his friend. (KJV)



Just following along this thread...no, I'm not trading actual money, and yes, I can see both RMI's and Q's points ! Day trade...swing trade..."trending"...whatever !!! :smile:

IMHO...this is Q's "system"...he has been MORE than open about when he feels the need for a change or an update to his system, and he's shared that for all to see...as well as his running results, either great or ugly ! I don't understand the underlying parameters of his system, but my feeling is...it's a work in progress ! That's why Q is sharing it with us all, instead of charging us 100's of dollars for his "system" !!!

Again...it's YOUR decision whether or not you follow his "system", whether by day or not !!! Just use it at your own risk...and I don't mean to imply that in any negative way... !!!


Stoplight...
 
Exactly, it is easier for me to do it 10 'till as I don't have to refresh a web page to get the call. I personally do it at the last minute to gain every advantage; however small.

All the crunching is done during the optimization process. There is a Long one that's done when I update the system; and there are daily short ones that I do after the close. So, it's just a matter of plugging in the current HIGHS/LOWS and reading the "instant" results of the calculations to generate the forecast.

However, I can do a second forecast as well. You decide: 15 or 20 minutes (or more?) before?

Q,

I agree with Stoplight. It is your system, you do whatever you think is right. Plus you have the advantage of knowing the ramifications of publishing your calls early vs late. With that said my thinking is most of the time the daily highs and lows are made throughout the day and not usually at the EOD. It would seem to me that statistically making a call 15 to 20 minutes before EOD would most likely have the same input data as at the EOD. However, I must admit I don't know how long it takes your system to "crunch" the data once it has been entered.



_______________________________________________________________
DISCLAIMER: Userque's trading signals are not affiliated with TSP Talk or Buy Low Sell High, Inc. The information is for educational purposes only! The information is not advice or recommendations. The information may be revised at anytime. Userque does not give investment advice. Do not act on this data. Do not buy, sell or trade any stocks, ETF's, indices, funds, or any other instruments/vehicles based on this information. Userque may trade differently than discussed or posted in this forum.
 
Many years ago, I found a website that claimed to use AI to generate a five day forecast (five days were free). Its forecast was actually a simple chart of the stock, extended ahead five days (plotted in a different color) based upon the forecasted prices. It was cool in and of itself, regardless of its accuracy (I don't remember how well it did...if I even knew that back then).

Another thing that's cool about such a forecast is that it doesn't give BUY or SELL signals. It gives a chart, and the viewer makes their own determination as to how to play the forecast.

The latest update to my system makes it possible for me to consider such a forecast for those following along in their demo accounts. I would still indicate specifically what my system is going to do however. I may work on that this weekend.
 
Last edited:
Optimization finally finished.

System would have remained in NUGT at close.

Next two days forecasted to be up days.

I'm going NUGT at open. I'll post a formal forecast in the a.m.




_______________________________________________________________
DISCLAIMER: Userque's trading signals are not affiliated with TSP Talk or Buy Low Sell High, Inc. The information is for educational purposes only! The information is not advice or recommendations. The information may be revised at anytime. Userque does not give investment advice. Do not act on this data. Do not buy, sell or trade any stocks, ETF's, indices, funds, or any other instruments/vehicles based on this information. Userque may trade differently than discussed or posted in this forum.
 
Last edited:
Disregard. There's something else I need to look into.

Optimization finally finished.

System would have remained in NUGT at close.

Next two days forecasted to be up days.

I'm going NUGT at open. I'll post a formal forecast in the a.m.




_______________________________________________________________
DISCLAIMER: Userque's trading signals are not affiliated with TSP Talk or Buy Low Sell High, Inc. The information is for educational purposes only! The information is not advice or recommendations. The information may be revised at anytime. Userque does not give investment advice. Do not act on this data. Do not buy, sell or trade any stocks, ETF's, indices, funds, or any other instruments/vehicles based on this information. Userque may trade differently than discussed or posted in this forum.
 
Maybe because its gapping up nearly 10%... sigh


?
:blink:

Why would premarket gapping cause me to say, "Disregard that the system....

Oh... nevermind, I see what you are suggesting.

I hadn't looked at any quotes this morning. And I never look at premarket trading as I don't trade premarket. I do look at futures, but I haven't this morning; been busy trying to get the system ready to trade the open and I'm in cash.

You may not be a programmer, so you may not realize that whenever you change one part of a program, sometimes, there are unintended consequences in another part of the code. Bad consequences. Hopefully, they are spotted and fixed.

I spotted one this morning.

For the record, I'm not one of those guys that looks at the futures or premarket action and concludes "Wow! today's going to be an up (or down) day!" Nor does my system look at futures or premarket action.

Moreover, my record speaks for itself with regard to being forthright and honest with regard to what I post and the results...however ugly or pretty. For scores of days now, I've forecasted the OPEN several hours before the OPEN, with no chance for reliance upon premarket action. Now... today, I'm suddenly moved by premarket action?

Now... today, I'm accused of child's play. Got it. :blink:

At least I did get a good laugh this morning. :)
 
FOR 10-22-15 AT CLOSE:
CASH

* Had to restart optimization again. No time to generate a forecast for the CLOSE.




_______________________________________________________________
DISCLAIMER: Userque's trading signals are not affiliated with TSP Talk or Buy Low Sell High, Inc. The information is for educational purposes only! The information is not advice or recommendations. The information may be revised at anytime. Userque does not give investment advice. Do not act on this data. Do not buy, sell or trade any stocks, ETF's, indices, funds, or any other instruments/vehicles based on this information. Userque may trade differently than discussed or posted in this forum.

When you update the plot of the results, which position will it reflect? Staying in NUGT or going to cash?


...
 
I wanted to get the optimization started prior to posting here and prior to investigating the bug.

I found that the original optimization was not affected by the bug. The bug would have only been a factor after today's close, when I would have updated the sheet with today's data. I'll repost the original forecast.

I was truly hoping the system would still have me going NUGT this morning. :nana:

_______________
"I laugh and I joke, but I don't play. I was kicked out of kindergarten for refusing to take recess."
 
FOR 10-23-15 AT OPEN:
ENTER NUGT




_______________________________________________________________
DISCLAIMER: Userque's trading signals are not affiliated with TSP Talk or Buy Low Sell High, Inc. The information is for educational purposes only! The information is not advice or recommendations. The information may be revised at anytime. Userque does not give investment advice. Do not act on this data. Do not buy, sell or trade any stocks, ETF's, indices, funds, or any other instruments/vehicles based on this information. Userque may trade differently than discussed or posted in this forum.

 
Done posting, apparently everyone thinks all I do is personally attack them...

Sorry you thought I was bashing your system or you, all I was saying was that if it changed maybe it was because NUGT was gapping up. I have no knowledge of the innards of your system, obviously.

Good Luck Mr. Q.
 
Done posting, apparently everyone thinks all I do is personally attack them...

Sorry you thought I was bashing your system or you, all I was saying was that if it changed maybe it was because NUGT was gapping up. I have no knowledge of the innards of your system, obviously.

Good Luck Mr. Q.

Thank you. You too.
 
Back
Top