The ACD Method
- Thread starter sbrowne126
- Start date
Watch that Euro and the 1.10 level...
Mav, you look at any of these numbers?
http://www.telegraph.co.uk/business...s-could-spell-disaster/?WT.mc_id=tmg_share_tw
http://www.telegraph.co.uk/business...s-could-spell-disaster/?WT.mc_id=tmg_share_tw
Nice work Bagger, great contributions to the thread. Out of curiosity, how did you structure your input data for this plot? For example, the NL daily score and 3 or five day NL will be the end result of the of the trade day (D). So would it be wise to adjust your data to where NL(D) is being compared to %Change(D+1) in order to analyze any predictive value? Thanks in advance.
I might be able to answer this for you.
The method you have described in your posts is the same way as I present mine whereby I have rows with nl scores in various day combinations and then previous returns along with future returns.
So then just by looking along one row I can see that when we had say a +10 what the return was nPeriods into the future. This makes subsetting and analysis a lot easier.
For this I use R's lead lag function, more about that here. In Python I believe that you can use bitwise operators such as << >> etc but I am not very Python savvy so someone else may be able to clarify.
Hope that helps.
The method you have described in your posts is the same way as I present mine whereby I have rows with nl scores in various day combinations and then previous returns along with future returns.
So then just by looking along one row I can see that when we had say a +10 what the return was nPeriods into the future. This makes subsetting and analysis a lot easier.
For this I use R's lead lag function, more about that here. In Python I believe that you can use bitwise operators such as << >> etc but I am not very Python savvy so someone else may be able to clarify.
Hope that helps.
That's pretty much what I do too. I will say I'm still assuming I made a mistake somewhere. I threw out that analysis but I remember the parameters and if I can recreate it with better tools I am building I will go print money. Again, I'm assuming a mistake...
Great article. Thanks for posting!
GC had a failure at the monthly A up today in conjunction with a daily C down. My 30 day went negative on 3/23 for the first time in two months.
Edit:
And some pretty sensitive NL derivatives I am playing with are just turning negative today and yesterday.
Edit:
And some pretty sensitive NL derivatives I am playing with are just turning negative today and yesterday.
Last edited:
GC as in gold - topped??
Doubt that...
I'm not a top caller lol. I've been mostly posting about tools, figure I should start talking markets instead. I only trade gold right now for a few reasons and I see a significant shift in gold. The new month and quarter are about to start so I think it's worth watching, and would be happy to take a short signal if I get one next week.