Trading fx fundamentals | Page 14 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Successful Fundamentals trading (DAY Trading WILL BE Ignored) 17 replies

Book suggestion for Forex News and Fundamentals Trading 1 reply

trading with fundamentals is the only way to go 129 replies

Trend Trading FUNdamentals 3 replies

importance of fundamentals/technicals in intraday trading 0 replies

Attachments: Trading fx fundamentals
Exit Attachments

Trading fx fundamentals

  • #261
  • Oct 27, 2020 3:52am Oct 27, 2020 3:52am
introducing the USD-IDX and AUD-IDX. I have spent a lot of time building the index logic and a GUI to allow me to rapidly explore data, build and configure indexes, and overlay them with currency data. I need to subscribe to an EOD exchange feed and have found one but have yet to integrate it. Next steps are to integrate the new FX feed into my index studio and the carry on building the indexes. Here are 2 images, the first is the USD-IDX overlayed with a historical dump of USD data. The second image is the USD-IDX overlayed with the AUD-IDX - this is the type of chart I will be using to trade from where I can see the divergences and have the directional bias. In the coming weeks, ill have the rest of the indexes built and integrated with exchange data and will begin testing trading and start looking at automation. I need to test this on a test account for a while before I move onto building a site and offering this as a service to others. The eventual goal is building indicators for these for MT4 and TradingView.
Attached Image (click to enlarge)
Click to Enlarge Name: usdidx.png Size: 75 KB
Attached Image (click to enlarge)
Click to Enlarge Name: usau.png Size: 40 KB
  • #262
  • Oct 29, 2020 3:57am Oct 29, 2020 3:57am
I admire your enthusiasm, but you say you want to make money from it. Maybe you should do a little market research before you start. I don't know about an AUD index but many others exist already (available for free):

DXY
EXY
JXY
CXY
and probably some more.
  • #263
  • Oct 29, 2020 4:15am Oct 29, 2020 4:15am
Quoting fxke62
I admire your enthusiasm, but you say you want to make money from it. Maybe you should do a little market research before you start. I don't know about an AUD index but many others exist already (available for free): DXY EXY JXY CXY and probably some more.
I've done the research, close to 1000 hours research to be more specific. Those indices you mention are comprised of a basket of currencies. I am not modeling exchange rate correlations, I am modeling economic data (no price inputs into the indices).
  • #264
  • Nov 1, 2020 10:16am Nov 1, 2020 10:16am
small update

I have started some RnD work on incorporating VIX. The goal is to fill the gaps between economic data and the standard deviation relative to the latest value of the VIX will serves as a weighing factor. I might need to parse news headlines to extract specific currencies and whether sentiment is positive or negative. I already have built the tech to do this for parsing central bank statements and it will serves as a means of targeting currencies with the weighting factor, thereby linking volatility with currencies and directional bias.

to give you an idea of how this might work with a risk currency like aud, here is the vix overlayed. Those spikes are areas where the index currently deviates correlation.
Attached Image (click to enlarge)
Click to Enlarge Name: IMG-20201101-WA0001.jpeg Size: 567 KB


by filling the gaps where sentiment can and does break economic data correlation, the indices should provide the feedback required to initiate or cut trades as well as make the overall index correlation more consistent and useful for trading.
  • #265
  • Nov 1, 2020 11:23am Nov 1, 2020 11:23am
Quoting EcoTrader
small update I have started some RnD work on incorporating VIX. The goal is to fill the gaps between economic data and the standard deviation relative to the latest value of the VIX will serves as a weighing factor. I might need to parse news headlines to extract specific currencies and whether sentiment is positive or negative. I already have built the tech to do this for parsing central bank statements and it will serves as a means of targeting currencies with the weighting factor, thereby linking volatility with currencies and directional bias....
What is the correlation or your general thesis here? As VIX or stock market uncertainty rises, then AUD/USD falls?
If you're gonna be a bear, be a grizzly
  • #266
  • Nov 1, 2020 1:20pm Nov 1, 2020 1:20pm
Quoting warrensapp
{quote} What is the correlation or your general thesis here? As VIX or stock market uncertainty rises, then AUD/USD falls?
Yes and no, it's not as black and white as stating when a plus b equals c then d. Most of the time when the vix rises there is demand for dollars. Risk currencies like aud tend to be hit but sometimes risk aversion leads to dollar selling so I am probably going to need to correlate vix with news flow to adjust what currencies are weighted and in which direction. My first pass however will be 20 years of data with static rules for risk currencies and separate static rules for the dollar.
  • #267
  • Nov 6, 2020 3:26pm Nov 6, 2020 3:26pm
Looks like an awesome project. I’m doing similar things but was a student with Ian so my focus is slightly different different.

If intraday gov bond data is being used for your analysis, where are you getting it? I have a gsheet
to pull the data from tradingeconomics and a macro to update it at various times but it’s a real hacky setup and I don’t like it.
  • #268
  • Nov 7, 2020 1:57am Nov 7, 2020 1:57am
Quoting rudy5
Looks like an awesome project. I’m doing similar things but was a student with Ian so my focus is slightly different different. If intraday gov bond data is being used for your analysis, where are you getting it? I have a gsheet to pull the data from tradingeconomics and a macro to update it at various times but it’s a real hacky setup and I don’t like it.
It's certainly keeping me busy. Regarding bonds, I don't use any bond data but for a good reason. I took various parts of the curve and overlayed them with the economy's ccy and its actually got a relatively weak correlation. It was one of the first things I looked at but was pretty disappointed in the weak correlation and therefore dropped it from being part of the indices. When I have the time I'll put together an example for you.

data isn't cheap, if you want intraday bonds you either have to pay hundreds per month up to thousands depending on your data requirements. I'm currently up to 300 a month with vendor data costs and this is cheap by industry standards. What are you building?
  • #269
  • Edited 2:32am Nov 7, 2020 2:17am | Edited 2:32am
Ok I'd been thinking the correlation between bonds and ccy's lately had been a let down so glad you agree.

A bot that will combine everything I know in order to derive a buy/sell rating for each currency.

1. price action, levels
2. economic data
3. options/ open interest
4. correlated markets

Which could then act as both a gate to allow some PA setups to be entered, and as a way to vary lot size. But also with the ability to also take a short term opportunity in an opposing direction to the longer term rating. Like say a CB interest rate announcement comes out and price drops, but then days/weeks later price trades above that level: subtract x from rating. Recent data has also been weak: subtract x. But then theres an upcoming random data release in the next 48 hours and price happens to be at some buy level: add x to rating, set to expire within n hours, allowing for lower time frame entries to be made against the overall sentiment.

I made an economic forecast gsheet a couple years ago using ff's cal which was ridiculously complex (ended up spreading it out across multiple sheets to not crash a browser). I've since upgraded to doing things in python but just recently ff started blocking requests. so back to the old gsheet, gotta scrape the scraper!

what vendors are you using? Tradingeconomics.com seems to be the gold mine but stupidly expensive
Attached Image
  • #270
  • Edited 5:20am Nov 7, 2020 4:06am | Edited 5:20am
I use trading economics, data exchange International, fxcm rates, and fred. Word of warning on economic data. If anyone knows me its one of the tools I use in trading but since this project started, I have seen how much of it is pointless for trading. Even indicators that's meant to be leading either lag or have little correlation. My focus these days has shifted more towards flows, sentiment, volatility as these offer far superior correlation. My view on this reflecting my time as a trader is the success I have had can be explained as sentiment trading. I would trade releases such as deviations and other related news. Now I know much of what I traded technically has lagging corrations or weak correlations, so price movement can only be explained as sentiment trading within the flows. My indices corrate price well but occasionally deviate. It is sentiment that's responsible for this and I now have the data to correct those and have consistent correlation. I am already at a point where I could start automating trading but putting it on hold until I manage to integrate a directional volatility weighting factor which will correct the index when these events happen.
  • #271
  • Nov 7, 2020 9:19am Nov 7, 2020 9:19am
What was your news trading strategy? Were you getting a directional bias based on the next few days forecasts, or were you actually trading the news?
  • #272
  • Nov 7, 2020 10:15am Nov 7, 2020 10:15am
Quoting rudy5
What was your news trading strategy? Were you getting a directional bias based on the next few days forecasts, or were you actually trading the news?

it still is a strategy I use very effectively. I don't look at single forecasts I look at the range of forecasts but not for directional bias. I trade when the actual is off the forecast and direction is based on analysis of news flow, sentiment, and central bank policy.

here is the 10y us gov bond going back 20 years - normal and inverted against the DXY


Attached Image (click to enlarge)
Click to Enlarge Name: b1.png Size: 82 KB
Attached Image (click to enlarge)
Click to Enlarge Name: b2.png Size: 83 KB


now compare that correlation with FX reserves - something nobody really talks about
Attached Image (click to enlarge)
Click to Enlarge Name: res.png Size: 80 KB
  • #273
  • Nov 12, 2020 12:10pm Nov 12, 2020 12:10pm
So I have been testing one of the indexes on a test account that's been active since 22 October. 41 trades, 1 loss so far and 103% gain on the account. It is still early days and will know more in the months to come. I am developing a new trading strategy around it and building an index studio with dashboards to trade off. Early days but promising start.
  • #274
  • Nov 12, 2020 9:15pm Nov 12, 2020 9:15pm
Quoting EcoTrader
{quote} I've done the research, close to 1000 hours research to be more specific. Those indices you mention are comprised of a basket of currencies. I am not modeling exchange rate correlations, I am modeling economic data (no price inputs into the indices).
Ah now I get it. Yes that is a very different beast.
  • #275
  • Nov 13, 2020 1:17pm Nov 13, 2020 1:17pm
scraping data is such a pain.... for some stuff (https://www.home.saxo/insights/tools...l/tool-details) all I would be able to do is take a screenshot and then use a python program to find the text inside the pic.
  • #276
  • Nov 14, 2020 2:32am Nov 14, 2020 2:32am
if anyone is in doubt about the correlation of oil and the Canadian dollar, similar correlations exist for AUD and NZ but for different commodities.

Attached Image (click to enlarge)
Click to Enlarge Name: canada.png Size: 138 KB
  • #277
  • Nov 14, 2020 10:10pm Nov 14, 2020 10:10pm
Cool, I posted something similar recently https://www.forexfactory.com/thread/...ude-oil?page=2
But note the big disconnect in 2018 / 2019.

With other commodities it might be (even) more reliable, because e.g iron ore is a more steady commodity.
Oil is dangerous/big opportunity for trading anyway (flip flops direction more than any currency pair except one) and
moves on average very fast (even faster than silver).

Interesting project you are working on - wish you good luck with it!
  • #278
  • Nov 15, 2020 2:50am Nov 15, 2020 2:50am
Quoting fxke62
Cool, I posted something similar recently https://www.forexfactory.com/thread/...ude-oil?page=2 But note the big disconnect in 2018 / 2019. With other commodities it might be (even) more reliable, because e.g iron ore is a more steady commodity. Oil is dangerous/big opportunity for trading anyway (flip flops direction more than any currency pair except one) and moves on average very fast (even faster than silver). Interesting project you are working on - wish you good luck with it!
2018-2019 can still be profitable trading the correlation. I don't tend to look for entire moves, I just don't trade that way. I find more opportunity and less holding risk by taking smaller bites. You see price zigzagging when that deviation happens. Well you can still trade the buys and sells conforming to the oil price direction and as you can see, many opportunities exists and slight break in correlation is ultimately due to fundamentals going on at the time (Our soon to be ex Dumpy Duck president of the USA).

just because correlation deviates doesn't mean opportunity is lost
  • #279
  • Nov 15, 2020 11:19am Nov 15, 2020 11:19am
If your beginner and struggling with market analysis are always reputable companies out there that can help and walk you through every step.
  • #280
  • Nov 15, 2020 12:05pm Nov 15, 2020 12:05pm
Quoting GlobalTForex
If your beginner and struggling with market analysis are always reputable companies out there that can help and walk you through every step.
Take a hike
0 traders viewing now