Sorry if this is a repeat topic. I tried the search function with no results. I've been using the RSI Wilder for some time, and I consider it one of the most useful studies yet, but recently, I've come across the RMI which smooths out the waves quite nicely. I leave the overbought and oversold values at 25 and 75, but I have the RSIW length at 14 and the RMI at 20 days. Does anyone have any experience with these who may have other tricks involving these studies?
Check out John Ehlers work on using digital signals processing (DSP) techniques applied to conventional indicators. Specifically, look at his use of the dominant cycle, inverse fisher transform as applied to the RSI. Generally, I prefer to use oscillators (like RSI) in an extreme value type of method. Remember, it is still an indicator and can only follow the event, not predict the event, which is price movement.
An indicator works or not based on if its settings are in sync with price movement. For example, sometimes using a setting of 14 will work perfectly to identify trends or tops or bottoms or whatever. But sometimes 14 might be exactly wrong, too, and 25 might work better. Or 17. Or 52. Or anything. And there's no way to know ahead of time. Nearly every (price based) indicator out there is only showing you what price is doing now relative to what it was doing a certain number of bars before. Put up RSI, MACD, stochastics, etc. on your chart and see how they pretty much all do the same thing. You may have to adjust the settings for each one a bit but you'll see what I mean. I've attached a chart showing this. They are all calculated differently but they're basically telling you if price is higher or lower than it was a certain number of bars ago. Some of them have calculations so that they flatten out as they approach a certain range and other ones will keep going up or down as long as price does, but they're all still doing the same thing.
Not quite. Different indicators tell you different things. Example: Stochs tell you where price is relative to a range. Macd simply tells you what a moving average is relative to another moving average. One indicator is superior to another. One lags a little, the other lags a heck of a lot. Also, one is measuring a range, the other is measuring momo.
Andrew Cardwell developed an indicator to mitigate this. In fact, his indicator would lead the RSI by a good bit, enabling the trader to react when his RSI is pegged or nearly pegged.
Which is the same as where it is relative to where it was a certain number of bars ago. Those moving averages are derived from price, so it's just a roundabout way of telling you where price is relative to where it was a certain number of bars ago. Even if one is superior to another, neither is useful. It's like saying getting punched in the stomach is superior to getting punched in the face. They're basically the same. Look at the chart. For the most part they both go up together, they both go down together. They both do the same thing.
Although I'm fairly new at this in the grand scheme of things, one of the first things learned was the absence of any crystal balls in the market. Understood. I know oscillators and indicators are just part of a larger system, and finding the settings that work for each individual can be tedious. There's a lot to learn and I'd much rather learn from other peoples' mistakes rather than doing it myself when it comes to this type of thing. So if anyone had any cautions about the RSIW or the RMI, I'm all about soaking up the info. I try not to put to much weight into them, but use them as an added tool to an overall convergence of evidence. Definitely in the realm of a confirmation, rather than an actual buy or sell indicator itself. Thanks for the responses. Any thoughts you might have as they pertain the RSIW or the RMI, please forward them.
It's a necessarily step for someone who is new to the market to do exactly what you are doing. You will spend weeks, months, or even years on this. You'll eventually decide RSIW and RMI are useless and you will move onto the next indicators, and you will repeat this process many times. If you stick with it, you will design your own indicators. And then you will eventually move on from them, too. We can tell you that there are no profitable indicators and there are no profitable systems that use indicators (the only people who disagree, btw, are people trying to sell you something), but you won't fully believe it until you go through this all for yourself. So keep doing what you are doing, because one of two things will happen: 1) you'll discover the holy grail and make a ton of money 2) you'll decide for yourself that indicators are not helpful in trading Here are some ideas to get you started: Since you'll eventually realize that following the normal buy and sell signals of your indicator will not result in net profitability, try reversing the signals (so buy when it says "sell" and vice versa). You'll see that the opposite of a losing system is still a losing system. Don't worry if it takes you a while to understand this. Take something like RSI, which has overbought and oversold levels, and watch how when price is in a sustained uptrend, it keeps going up even though RSI says overbought. Of course, you had no way of knowing beforehand that price was going to be in a long uptrend. RSI just tells you what price is doing relative to what it was doing before, and then someone decided "ok, above here is overbought and below here is oversold.' Those levels have nothing to do with anything. RSI just tells you what price is doing now relative to what it was doing before.