DJIA News and Forecasts

[Yomiuri]

Neural Network - an artificial intelligence. It analyzes the emerging news, making fundamental and technical analysis. In this issue, we look at the stock index futures on the DJIA. On average, the stock market is growing at 10% per year. So I open and hold only the bullish position. Neural network recognizes the negative factors in the economy and gives the signal to withdraw from the market. Forecasts published about 2pm-3pm New York time. Frequency forecasts 1-2 times a week.

Forecasts: open long position

 

[qldfrog]

Re: Neural Network DJIA

I will follow you with interest, NN was one of my specialisation when i graduated.Always thought about giving it a go with stock markets
All the best

 

[Wysiwyg]

Re: Neural Network DJIA

Neural network recognizes the negative factors in the economy and gives the signal to withdraw from the market.

Hello. What would your inputs be? Twitter and Facebook feeds, GDP - Non Farm Payrolls - ISM - PMI - Interest Rates, etcetera. What T.A. do you use? What is your publication? Be an interesting thread if you want to elaborate on the process used to conclude on market participation.

 

[Yomiuri]

Re: Neural Network DJIA

Hello. What would your inputs be? Twitter and Facebook feeds, GDP - Non Farm Payrolls - ISM - PMI - Interest Rates, etcetera. What T.A. do you use? What is your publication? Be an interesting thread if you want to elaborate on the process used to conclude on market participation.

Neural network uses a huge amount of information. All known methods of fundamental and technical analysis. Since the neural network is very complicated, it is difficult to explain how it works. Just watch the forecasts. I think that this topic will help you in your trading.

 

[Wysiwyg]

Re: Neural Network DJIA

Just watch the forecasts.

What is that? Open long position on the Dow as per your first post?

 

[howardbandy]

Re: Neural Network DJIA

Hi Yomiuri --

It will help everyone watching your postings to know some of the characteristics of your predictions. Such as:
How often is the model run? Daily? Weekly?
What outputs can it have? Long? Flat? Short?
What time of day, relative to the open or close of trading, is the data gathered?
How long after the data is gathered is the signal generated?
How long after the signal is generated is it published to this forum?
How regular will your posting be?
And anything else that you care to disclose.

Best regards,
Howard

 

[Yomiuri]

Re: Neural Network DJIA

How often is the model run? Daily? Weekly? - Daily, Weekly
What outputs can it have? Long? Flat? Short? - Only long
What time of day, relative to the open or close of trading, is the data gathered? - all time
How long after the data is gathered is the signal generated? - 2pm-3pn New York time
How long after the signal is generated is it published to this forum? - 2pm-3pn New York time
How regular will your posting be? 1-2 times a week

 

[Gringotts Bank]

Re: Neural Network DJIA

How often is the model run? Daily? Weekly? - Daily, Weekly
What outputs can it have? Long? Flat? Short? - Only long
What time of day, relative to the open or close of trading, is the data gathered? - all time
How long after the data is gathered is the signal generated? - 2pm-3pn New York time
How long after the signal is generated is it published to this forum? - 2pm-3pn New York time
How regular will your posting be? 1-2 times a week

What software or subscription service are you using to generate the NN signals?

What sort of win rate have you been getting?

cheers

 

[Wysiwyg]

Re: Neural Network DJIA

What is that? Open long position on the Dow as per your first post?

Still long the Dow? Is there a sell signal?

 

[Yomiuri]

Re: Neural Network DJIA

What software or subscription service are you using to generate the NN signals?

What sort of win rate have you been getting?

cheers

The basis I took NN the same big bank, I redid it. Early it worked with currency, now with the stock market.

10% per year on the index and a leverage of 1:3. I have an income of 30% a year.

 

[Yomiuri]

Re: Neural Network DJIA

Still long the Dow? Is there a sell signal?

Still long

 

[Wysiwyg]

Re: Neural Network DJIA

Still long

Okay since no entry price was given I will give you best price of daily low on 18/05/2016 at 17418 'cause I'm a nice guy. I mean what's a 200 point range between friends, right.

Enter Dow long = 17418

 

[Ironik]

Re: Neural Network DJIA

Neural network uses a huge amount of information. All known methods of fundamental and technical analysis. Since the neural network is very complicated, it is difficult to explain how it works. Just watch the forecasts. I think that this topic will help you in your trading.

Hi Yomiuri,
I'm watching this space with some interest but it looks like you are not willing to educate us.
Please share more details about this:All known methods of fundamental and technical analysis. Since the neural network is very complicated, it is difficult to explain how it works.
Please remember there are a lot of noobs like me who are going to read this and try to make some sense of what you are on about.

Thank you very kindly.

 

[ThingyMajiggy]

Re: Neural Network DJIA

Riiiighhht....

Comes on ASF about the latest rage in trading in neural networking/machine learning, is about as brief as passing through an outback town, no specifics whatsoever, no entry prices, no explanations, just a "forecast 1-2 a week". What's the point? No one cares how big your gentleman sausage is, so just get on with explaining(or trying to, we get it's complicated, so is trying to get anything out of your current answers) what the go is and some details, or at the very least answer the questions that others have asked.

Who coded it? What's the timeframe? What platform and language is it used in/on? How does it generate it's signals, and what are they based on?

You do know this is a forum right? As in discussion and stuff

 

[Yomiuri]

Re: Neural Network DJIA

Riiiighhht....

Comes on ASF about the latest rage in trading in neural networking/machine learning, is about as brief as passing through an outback town, no specifics whatsoever, no entry prices, no explanations, just a "forecast 1-2 a week". What's the point? No one cares how big your gentleman sausage is, so just get on with explaining(or trying to, we get it's complicated, so is trying to get anything out of your current answers) what the go is and some details, or at the very least answer the questions that others have asked.

Who coded it? What's the timeframe? What platform and language is it used in/on? How does it generate it's signals, and what are they based on?

You do know this is a forum right? As in discussion and stuff

What do you want? Exchange trading is very difficult to do. In this work, a lot of unknown. This is one of the most difficult jobs in the world. You're trying to get closer to the truth. I understand you. In the past, I myself was the same.

 

[ThingyMajiggy]

Re: Neural Network DJIA

What do you want? Exchange trading is very difficult to do. In this work, a lot of unknown. This is one of the most difficult jobs in the world. You're trying to get closer to the truth. I understand you. In the past, I myself was the same.

Nevermind, just a bit of detail and elaboration and those that have asked questions to be answered. But I believe I am flogging a dead horse, shall be watching with interest

 

[qldfrog]

Re: Neural Network DJIA

To come to the help of the poster:
To explain how neural network works is not easy;
in a nutshell, imagine an articial brain as a black box;
you feed it various scenarios and outcomes from the past; heaps of them=> so basically historical data (to be defined)
and teach that brain to detect a positive outcome; after a while , the brain will tend to recognise patterns which should end up positive, or not;

Once trained, an ideal NN for each past scenario should tell you:
in that case, the day after, the ASX will be at xxx or the SP will reach , etc;
you then expect to feed live data to the "brain" and the brain telling you then what is (the most likely) to happen in that case;
Hopefully, this should give you an edge.
BUT
If the brain tells you: "you need to buy XX today and you will profit tomorrow", at no time can you (or the brain) explain why there is such an assumption.
What is important and can restrict the model is:
how wide is your data feed for scenario, do you include weather? Do you include news on the latest bomb in Paris, usually you do not, yet it is obvious this may affect the market.
And how do you restrict your domain?
for example: if my NN just care about AAC share price: Iinclude weather, AUD/US, ASX move...but are you going to fed the latest row between Indonesia and Australia?Will you detect the Mouth and Foot disease alert?
So even restricting the universe may not be enough to remove sizable risk from the prediction;
I hope this helps?

It is very interesting; i have projects right and left, startup work, innovative apps, and a life, but if one day I ever get bored, a NN in amibroker is in my TODO list

 

[Ironik]

Re: Neural Network DJIA

Hi qldfrog,
Thanks for taking the time to explain this.
I really appreciate it. It looks like this would be out of my league and do not intend to pursue this further. Just thinking about the input parameters to NN, it seems close to infinity.
I do use Amibroker and hope that one day they come up with something close to NN capabilities.

Thanks again for your time.

Ironik

 

[qldfrog]

Re: Neural Network DJIA

Hi qldfrog,
Thanks for taking the time to explain this.
I really appreciate it. It looks like this would be out of my league and do not intend to pursue this further. Just thinking about the input parameters to NN, it seems close to infinity.

that is why we will NEVER have a real 100% fool proof model for NN nor will we even have truly precise weather forecasts

I do use Amibroker and hope that one day they come up with something close to NN capabilities.

Thanks again for your time.

Ironik

hum would take a while to develop the engine, and then you are only starting
the danger I see is that i do believe NN can definitively increase your odds, and if it does the only want to increase your return is top do more trades with more money;
let's say you are JO Morgan and /or all the big players in the world;
here you go shuffling billions and billions (they are probably already doing this) in a fully automated light speed way, and then $hit happens, whatever it may be and within 20 seconds you might have put billions on sell, triggered stop loss among international markets and created a minicrash until your cocaine sniffing agent can press on the red button;

i do believe many of these mini crashes are actually generated by these systems either via bugs or just extreme but sensible reaction.Once to many of these will start feeding into the NN learning processes , we might see some nice feeding loop crazyness happens
Anyway , Yomiuri does my "explanation" made sense?

 

[ThingyMajiggy]

Re: Neural Network DJIA

To come to the help of the poster:
To explain how neural network works is not easy;
in a nutshell, imagine an articial brain as a black box;
you feed it various scenarios and outcomes from the past; heaps of them=> so basically historical data (to be defined)
and teach that brain to detect a positive outcome; after a while , the brain will tend to recognise patterns which should end up positive, or not;

Once trained, an ideal NN for each past scenario should tell you:
in that case, the day after, the ASX will be at xxx or the SP will reach , etc;
you then expect to feed live data to the "brain" and the brain telling you then what is (the most likely) to happen in that case;
Hopefully, this should give you an edge.
BUT
If the brain tells you: "you need to buy XX today and you will profit tomorrow", at no time can you (or the brain) explain why there is such an assumption.
What is important and can restrict the model is:
how wide is your data feed for scenario, do you include weather? Do you include news on the latest bomb in Paris, usually you do not, yet it is obvious this may affect the market.
And how do you restrict your domain?
for example: if my NN just care about AAC share price: Iinclude weather, AUD/US, ASX move...but are you going to fed the latest row between Indonesia and Australia?Will you detect the Mouth and Foot disease alert?
So even restricting the universe may not be enough to remove sizable risk from the prediction;
I hope this helps?

It is very interesting; i have projects right and left, startup work, innovative apps, and a life, but if one day I ever get bored, a NN in amibroker is in my TODO list

If you're using the history of the instrument to train the ML component, wouldn't that then include all you mentioned, because we've had crashes, random announcements, terrorist attacks, all of that has happened before so the volatility is there and being included. All any of us have is what the instrument has done in the past anyway, all of those things you mentioned can catch any one of us out whether it's just us trading it or if we're using ML to assist.

I definitely think it has potential to help, at least in a direction manner at the very least, highly likely that tomorrow will be up or down, not necessarily an exact "it's going to this number" etc. Even if it predicts with reasonable accuracy what direction we're headed it would still be hugely beneficial. I'm currently going through the process of making one with python based on the DAX so I'll see if there's anything in it.

Definitely agree on these crashes being systems feeding on each other until someone kicks the plug out of the wall before we hit the -1000 point mark, the whole 2006(or whenever it was) flash crash being a fat finger trade by the one trader they pinned it on is the biggest lot of rubbish I've ever heard.

But it would be nice to hear a little more about Yomiuri's, if he's to the point of actually using/trading with it in the real world