This is also aimed at macd, RSI, stochastics, etc.
Can someone explain to me why using an indicator that is a derivative be it a linear or non-linear transformation of price can be used to confirm the movement of a price?
Confluence of indicators i agree with but surely the system is massively colinear?
Hello Pipwhip! I cannot answer your question but I will cheekily suggest that this article may be a good stimulus to interpreting RSI signals the ārightā way:An Introduction To The Relative Strength Index. Question to you: do you use RSI/oscillators in conjunction with other signals/indicators, and if so, which ones? Cheers.
Thanks for the info. I know the classical reasons people give about divergence etc. I was really looking for an answer in relation to information theory and projections of variables through non-linear functions. Iām going to have to do some reading.
I use an ema and volume. Interestingly i would say you could make a similar point about volume. So maybe if there is true value in volume, maybe there is value in RSI.
I was hoping someone could shed some light in regards to correlation being different and two functions that are non linear projections of the same underlying variable have information content when the result of the outputs diverge. Iām probably
RSI is a momentum oscillator that is able to measure the velocity and magnitude of stock.
Now, be honest, did you actually read the question before anwsering?
Iām not going be able to use the mathy speak that you do in regards to linear/non-linear, but I will say this ā the only place Iāve been to that really preaches how markets are in fact non-linear and our approach to it should be as such is fxgroundworks.com, which is a site based on harmonic trading believe it or not, but also has non-linear indicators based on sonic waves and things like that. Damned if they donāt work really well on longer timeframes, I was impressed.
And FWIW, I think the RSI is relatively worthless, but the system I use is a confluence of about 5 linear indicators with the settings I like, and it works fantastic. Iām not one that likes to share (Iāve lost too much money trying everything else until I finally found the combo that works ā nobody is getting a free ride out of me at this point), but one of them I will let out is the TSI indicator, which is much more smoothed, and 100X more effective than the RSI in my opinion. Everyone using RSI should switch to this immediately.
Sorry, but posts like this make me laugh a little.
Which bit? The rsi being worthless?
Baseless snark, nice job!
There are die-hard price action people who turn their noses up at almost every indicator, and thatās fine, you can make money a number of ways, and what they like/dislike doesnāt affect me any. But if you do use RSI, I canāt think of any reason to not use something thatās more smoothed out and less misleading like the TSI. Itās a necessary component of my overall system, and Iām a big fan.
Trust me, Indicators are put ther to suck your money much much faster. DItch all that crap and make your own work.
I should have probably posed my question in a slightly different way. I have at least managed to make a bit more sense of what i was trying to ask. It was effectively how can a derivative of a time series be used to forecast the original price series.
Before i give my reasoning i would just like to say i donāt give a sh1t about any single indicators in particular.
Ok, so I think the answer is yes you can forecast a time series with a derivative of itself. My reasoning is because its possible in some time series to find first order auto correlation. This is the simplest case where a time series can forecast itself. Using this logic there is no reason that a more complicated non linear transformation could not be also useful in forecasting.
case closed, thank you for your participation.
???
Dude, it is everything ok with you? Nobody can predict price movements, nobody and a math formula cannot replace the hard work.
Is your opinion that it can never be predicted or that itās cannot be predicted all the time?
Just a couple of things:
You may want to make your question less technical. Youāre limiting your audience to the people that understand what youāre saying. I canāt really follow it.
Anytime somebody comes on here and speaks definitively, itās best to ignore them. There are no real definites in Forex.
The whole reason I made the move from stocks to Forex is because I was curious to see what this currencies thing was that everyone in the stock forums kept saying was āA good way to lose all of your moneyā. After reading this over and over I had to ask myself, āOk, so how much real time and effort did these people put in? Could they even read a chart? How many times did they allow themselves to fail before giving up? Once? Probably.ā
The same applies to all of the people I saw speak ill of indicators. Did they try a lot of them? Did they try combinations of them? Did they backtest and forward test them? Are they just regurgitating what other people were saying? Chances are, just like the stock traders that were poo-pooing Forex, the price-action zealots almost certainly gave up on indicators too soon. Thatās fine, they can squawk away.
Apart from S&R, which I only really use to scale out, Iām all indicators, and itās been really great for over a year now once I got the combination I liked.
Bottom line, keep an open mind and find what jibes with your style. Ignore the people out there that tell you that x always = y. And dumb down your queries a bit so we can help you better. 
Oh no, iām sure itās not worthless, depending how it is used. The funny part is when, and not just limited to this post as iāve seen it many times before, a person suggests using a tool over another tool a certain way without any proof that it is indeed more efficient this way. They then go onto say āitās way better, increases win rate, everyone should use itā¦blah blah blahā
It seems people forget that the Forex market goes only two waysā¦ yet there are thousands of tools out there. There is certainly no cut and dry ābetterā tool, as the main variable will be the trader using it! As with most things, it comes down to what the individual in question finds to be the best supplement for their trading.
This is a question that has perplexed me for some time, not from a math perspective, but from a logical viewpoint.
Is it logical to predict the future action of A based on the past action of A? - or is it better to introduce the actions of B that have an influence on A?.
Larry Williams argues that the former is illogical, he thus searches for the latter.
Today, for example, there was much excitement in Fibre, many retail traders were short or were tempted to go short at around 3800/20, with maybe a stop at slightly above 3850.
Iām guessing that RSI and many other indicators were telling them to go short here - oversold etc.( A on A)
I wonder if maybe before going short that a look at Gold, S&P futures, US10yr and US30yr ( A on B ) would have suggested that the move was illogical, or whatever.
I post this simply because I too was sorely tempted to go short as above, I didnāt because of B.
Could I add to the above list, when it comes to the Euro, I now include the DAX, often called the GER30 on MT4.
Thanks to a certain poster who used to be on FF, sometimes also here, who I suspect has a certain dislike for scammers and such like, he is a little like Zoro, hard to track down but you learn a lot when you encounter him.
I donāt know, maybe simply opening the weekly EURUSD chart is all youād need.
Seems like a quick look at the previous weekly candles would clearly signal going short EURUSD at 1.3800/20 or 1.3700/20 would be like standing in front of a high-speed freight train. :56:
Do you really need to complicate things by watching Gold, S&P futures, US10yr, US30yr, etc to figure out EURUSD when the weekly candles are clearly telegraphing long, long, and long?