Williams Vix Fix BB + RVI + LinReg & Squeeze (Keltner) BBW + %BLegend:
Entery signal: When line color turns to lime (lighter green) after a blue dot appears
Exit signal: When line color turns to red (darker red) after a red dot appears
Note: it is more affective as an entry signal (Bottom is stronger)
- When line touches or crosses red band it is Top signal (Williams Vix Fix)
- When line touches or crosses blue band it is Bottom signal (Williams Vix Fix)
- Red dot at the top of indicator is a Top signal (Relative Volatility Index)
- Blue dot at the top of indicator is a Bottom signal (Relative Volatility Index)
- Gray dot at the bottom of indicator is a Keltner Squeeze signal (filtered by either BBW or %B)
- Silver dot at the bottom of indicator is a weaker Keltner Squeeze signal (Doesn't meet either BBW or %B filter)
- Purple is a 'Half Squeeze' only 1 Bollinger Band crossed the Keltner Channel
This is an attempt to make use of the main features of all 6 of these Volatility tools:
- Williams Vix Fix + Bollinger Bands
- Relative Volatility Index (RVI)
- Linear Regression (detects Vix Fix starts to rise or fall to a certain degree in order to help validate bottom/top)
Note : There is also added precision on Linear Regression entry by dividing WVF by square roots of basis.
- The crossing of Keltner Channel by the Bollinger Bands (Squeeze)
Conditions to Help Filter Keltner Squeeze:
- When the Bollinger Bands Width (BBW) value is lower than the lowest value within a period plus a margin of error (percentage)
- When the %B value goes up or down by the impulse value (threshold value in setting) detailed in LazyBears indicator. (www.tradingview.com)
If it meets one of these 2 filters and there is a Keltner Channel Squeeze than gray color or else if the squeeze doesn’t meet one of the 2 filters than silver color (weaker Squeeze).
The goal is to find the best tool to find bottoms and top relative to volatility and filter squeeze.
Note: You can also change the threshold for RVI top and bottom.
And this work builds on my last indicators:
- Williams Vix Fix + BB & RVI (Top/Bottom) & Squeeze ()
- Williams Vix Fix BB + RVI & Squeeze (Keltner) filtered BBW + %B ()
If you have ideas on this work or have ideas on potential combinations please message me, I always want to learn or get perspective on how it can be improved.
Sharing is how we get better (Parameter tuning, ideas, discussion)
I don’t reinvent the wheel, just trying to make the wheel better.
Keltner Channels (KC)
Multiband Oscillator - Zigzag versionJust variation of Multi Band oscillator present here: Multi-Band-Channel-Oversold-Overbought-Oscillator
Changes are:
Instead of regular moving average, here I am using Zigzag Moving average. This is calculated in similar to as explained in: Zigzag-Cloud
Instead of ATR, using AZR (Average Zigzag Range) - Average-Zigzag-Range-AZR
Rest of the logic remains same.
Number of bands used 100 - which means, calculate 100 Bollinger bands with Std Dev Multiplier starting from 0.1 and with step 0.1 for the next one.
Which divides price ranges into 200 equal parts. Calculate what is the current range and plot them.
Overbought - Oversold levels are dynamics. They are dependent on the max and min state price has reached in last 80 days. Offset and factor can be used to adjust overbought oversold levels.
Keltner Worm ForecastBeta V0.1.0
This Keltner Channel Worm gives 6 different KCs at max, and allows you to set a multiplier for the offset and worm settings (exponential).
The idea behind this script is that the overall brightness of the Keltner Channel will indicate the strength of the forecast. If the candles are shrouded by darkness, a correction is due.
Enjoy and let me know if you have any suggestions for improvements down below!
<3 Happy Trading
Sir Poggy
Dual Mean Reversion Channel (adjusted lower band)This is a public and open-source lighter version compared to the "Overextended Price Channel" which is provided complimentaty to the Trend Insight System.
Introduction :
Channels are very useful tools to assess overextended price, volatility and upcoming retracement or impulsive moves (such as Bollinger Band squeezes). It is an indispensable addition to any trader using Mean Reversion theory for a scalp-trade or swing-trade.
This script contains :
- 2 channels Keltner-style, using the True Range for volatility
- customizable volatility (channel width) and smoothing period
- a standard selection of moving average ; SMA, EMA, VWMA
- an embedded readjustment of the lower bands to avoid the drop on a logarithmic scale (see explanation below)
Why another channel indicator ?
I have found most conventional channels to be either not based on "proper" volatility (e.g. standard deviation of price action for Bollinger Band), or the bottom channel to be ill adapted to the logarithmic scale and plunges to 0 on some high volatility periods, messing with readability on logarithmic auto-scaled chart.
Also, I find the channels to be most useful when superimposed with another one of longer length; especially a pair of channels with a 50 and 200 period moving average respectively. Mean Reversion traders that mostly trade the 50 and 200 SMA/EMA know what I am talking about as having a channel helps to have a better visual for a proper of entry and exit point.
Disclaimer :
This indicator was originally intended to be used along with the Trend Insight System to improve performance, and the default configuration mostly backtested on BTCUSD.
Please use with caution, proper risk management and along with your favorite oscillator, candlestick reading and signals system.
Some explanation :
Based on Mean Reversion paradigm, everything has a tendency to revert back to the mean :
- when the price enters the upper channel, it is supposed to be (or start getting) overbought as the market is getting overheated, thus prone to correction,
- on the other hand, when the price enters the lower channel, it is supposed to be (or getting) oversold and the market looks favorable for a buy-in.
Depending on the trading style used, a trader will usually either wait until the price leaves the channel towards the mean before taking action (conservative style) or you will set limit orders inside the channel as you expect a reversion to the mean (more agressive/risky style).
With two channels, more complex (and maybe precise) rules can be built to optimize one's trading strategy.
Important notes :
In the end, sticking with 50/200 length and a single setting on volatility might be wiser, be wary of overoptimization which is risky at best and counter productive at worst (according to legendary traders such as Mark Douglas). Even if, needless to say, the volatility needs to be adjusted between a nascent and volatile market (such as crypto) compared to standard call markets that are much less volatile.
End notes :
It will always be considered a work in progress to help bring out the best of trading with channels, any comment and suggestion are welcomed.
Squeeze Momentum [Plus]The "Momentum" in this indicator is smoothed out using linear regression. The Momentum is what is displayed on the indicator as a histogram, its purpose is obvious (to show momentum).
What is a Squeeze? A squeeze occurs when Bollinger Bands tighten up enough to slip inside of Keltner Channels .
This is interpreted as price is compressing and building up energy before releasing it and making a big move.
Traditionally, John Carter's version uses 20 period SMAs as the basis lines on both the BB and the KC.
In this version, I've given the freedom to change this and try out different types of moving averages.
The original squeeze indicator had only one Squeeze setting, though this new one has three.
The gray dot Squeeze, call it a "low squeeze" or an "early squeeze" - this is the easiest Squeeze to form based on its settings.
The orange dot Squeeze is the original from the first Squeeze indicator.
And finally, the yellow dot squeeze, call it a "high squeeze" or "power squeeze" - is the most difficult to form and suggests price is under extreme levels of compression.
Now to explain the parameters:
Squeeze Input - This is just the source for the Squeeze to use, default value is closing price.
Length - This is the length of time used to calculate the Bollinger Bands and Keltner Channels .
Bollinger Bands Calculation Type - Selects the type of moving average used to create the Bollinger Bands .
Keltner Channel Calculation Type - Selects the type of moving average used to create the Keltner Channel.
Color Format - you to choose one of 5 different color schemes.
Draw Divergence - Self explanatory here, this will auto-draw divergence on the indicator.
Gray Background for Dark Mode - to make them more visually appealing.
Added ADX (Average Directional Index) that measure a trend’s strength. The higher the ADX value, the stronger the trend. The ADX line is white when it has a positive slope, otherwise it is gray. When the ADX has a very large dispersion with respect to the momentum histogram, increase the scale number.
Added "H (Hull Moving Average) Signal". Hull is a extremely responsive and smooth moving average created by Alan Hull in 2005. Have option to chose between 3 Hull variations.
Added "Williams Vix Fix" signal. The Vix is one of the most reliable indicators in history for finding market bottoms. The Williams Vix Fix is simply a code from Larry Williams creating almost identical results for creating the same ability the Vix has to all assets.
The VIX has always been much better at signaling bottoms than tops. Simple reason is when market falls retail traders panic and increase volatility, and professionals come in and capitalize on the situation. At market tops there is no one panicking... just liquidity drying up.
The FE green triangles are "Filtered Entries"
The AE green triangles are "Aggressive Filtered Entries"
(JS) Interchanging ATR & VWAP BandsOkay so this is pretty simple, but I think it's a great tool for day trading especially. I just took the default VWAP and Keltner channel scripts and combined them together.
The top option allows you to choose which one you'd prefer to use, "Use ATR instead of VWAP" .
The next options, "ATR Source", "KC Length", and "ATR Length" are the parameters for the ATR Bands.
"Number of Bands" allows you to choose how many bands you'd like to be on display (you can choose 1-8).
"Use Expoential MA" and "Band Style" are more default parameters from Keltner Channels used to set up the ATR Bands.
The "ATR Bands" are just stacked Keltner Channels separated by 1 ATR each, whereas the "VWAP Bands" are separated by standard deviation just like the default script from Trading View.
In these example chart, you can see the weekly VWAP with 8 deviation bands and 5 ATR bands with Keltner Channels.
Peak Reversal v2This is a brand new version of my Peak Reversal indicator. As with the older version, the idea behind this indicator is simple: identify potential price reversal areas, and identifying markets which are trending. In this new version I focused on improving on the old concept, but introduced a bunch of features heavily inspired by Adam Grimes' ideas from The Art and Science of Trading. (I also blatantly stole the way he colors candles outside of the bands. Sorry.)
As you can see below this indicator gives traders a plethora of tools to judge whether a market is trending, and might be mean reverting soon.
Follow me, join my group, like the script. You know the drill.
Basic functions:
You have a triplet of Keltner (ATR-based) bands in Peak Reversal. They are defined by a multiplier and an EMA, which is referred to as "the mean". There's a tight, normal, and an extreme band. The multiplier defines how far apart your bands are. By default the indicator uses 1.125, 2.25, and 3.375. The tight band is off by default, but you can turn it on in the options. The mean is also off by default. This is more a personal preference thing for me, because I happen to use a different indicator to show a couple of moving averages.
Band crosses:
Peak Reversal can indicate whenever price crosses one of the bands. This can help traders identify points where a mean reversal play could be an option. Triangles indicate these crosses. New in version 2 is the ability to choose which of the bands to use to show these crosses. If you are more of an aggressive trader, you might find it better to show tight band crosses. If you are looking for more extreme market conditions, then choose extreme. The default is "normal".
Free bars:
Indicating free bars is also a concept from the book. A "free bar" is one which stands "freely" above the bands, which means its low price is completely outside of the bands. It can be argued that a freely standing bar is an even more extreme mean deviation, than just a band cross. Traders can gain an additional advantage studying the markets this way. Free bars are not shown by default, when on, a star shape on the candles indicates free bars. Both band crosses and free bars can be shown at the same time, but there might be overlap.
Deviations:
Also based on a concept from The Art and Science of Trading, is an indication of price "deviations". You will notice that when a candle "touches" a band (high and close above band), its colored. The idea here is to show traders when a market is in motion, but also when a mean reversal might be coming next. To accomplish this, the more colors deviate, the darker the color is. The idea here is also simple, the more price deviates off the mean, the likelier it is to return to it. This uses three different shades to show these deviations. 1-2 is one shade, 3-4 another, and upwards of 5 there's only the darkest shade. I didn't make extensive studies, which color for how many candles would be appropriate to use, but I do believe it doesn't matter that much in usage. It's clear what traders gain from using this information: more deviation, the likelier a snapback becomes.
Advanced mode:
Last but not least, I decided to add an advanced mode for advanced traders. This does nothing more than flip all colors and shapes upside down. Everything that is red, becomes green. The idea is where some traders say "buy low, sell high" (standard mode), other traders might say "buy high, sell higher" (advanced mode). See for yourself, which one you like better.
Inverse BandsThis was the result of quite some time spent examining how much information could be gleamed by studying the interactions between Keltner Channels, STARC Bands and Bollinger Bands. I was surprised by the results.
First of all, there are four fills that are black. Set the transparency of those to 0 and you'll see this indicator the way that it's meant to be seen. Those fills belong to unused sections of the Bollinger Bands.
There are two clouds which represent STARC Bands and the Keltner Channel. There is some delay when they flip from bullish (green) to bearish (red), but they are indicative of the trend. The space between them is black and the narrower that space is, the greater volatility is. Because of this, we don't need the exterior Bollinger Bands.
The Bollinger Bands remain visible as the yellow interior clouds on the top cloud and the blue interior clouds on the bottom cloud. Often, the thicker the yellow or blue cloud is, the less severe a throwback from a given trend reversal will be. Often the thinner that yellow or blue cloud is, the more severe the trend reversal will be. If price is rising into a thin interior yellow cloud, the following dip will be substantial. If price action dips towards a thicker interior blue cloud, often the pump following that dump will be less enthusiastic.
We preserve the Keltner Channel and STARC bands as our cloud because the way that they interact with the three basis lines yields a lot of information.
The yellow Bollinger basis line tells us about trend strength. The closer the BB basis line is to the top of the top cloud or the bottom of the bottom cloud, the stronger the trend is. When it enters the cloud very close to the bottom of the bottom cloud, you know you're looking at a strong pump, and vice versa when it's close to the top of the top cloud.
The purple Keltner Channel basis line and orange STARC Band basis line can forecast short term trend changes one candlestick in advance by contacting any line in either cloud. The moment either basis line touches or crosses any boundary of the clouds, you know that the next candle will change directions. In an uptrend, a touch or cross means the next candle will have a lower high point. In a downtrend, a cross or touch means the next candle will have a higher high point. This is most useful in scalping.
It'd be pretty easy to slap some crossover alerts on to this and useful considering that they come a candle in advance. Feel free to further explore and develop this.
Keltner FibzonesKeltner Channel with Fibonacci Zones which uses properties of a Donchian Channel.
This script is a variation of the Fibonacci Zone script and the Donchian Fibonacci Trading Tool which I published earlier. A Keltner Channel gives more useful information to the trader than a Donchian Channel, because it provides a depiction of “normal range” and shows “outside normalcy” situations. Fibonacci lines provide a way to see if the market is trending up or down, while moving inside the channel, because in cases where the Fibonacci lines function as resistance, the trend is down, whereas when these turn out to be supports the trend is up.
Example of use:
If after a rally up - in which candles moves outside the channel - these return into the channel, this means a “new normal”. If the fibs turn out to give support, one may assume that this is a temporary sideways movement in which a flag is formed, after which following rally up may happen. With rally down the opposite is true. Top and bottom situations show a change in the way the market uses the fiblines.
The middle line of the Donchian Channel is used as the middle line of the Keltner channel, in stead of the sma in the classic channel. Default for calculation is 2 x Average True Range above and below this line. Default for the periods of the channel is 20 periods, because this allows the candles to go outside the channel. If you shorten this, all happens inside the channel.
Squeeze Momentum Indicator v4_pine [By Lazy Bear]This is the famous Squeeze Momentum Indicator made by @LazyBear in v4 version if someone wants to test some strategies, as the original code was in v2 version the code converter couldn't convert to v4.
Built-in Kelly ratio for dynamic position sizingThis is the defaut keltners channel strategy with a few additions.
The main purpose is to show how we include the Kelly ratio into our scripts for dynamic position sizing based on the performance of the strategy on a per trade basis.
We've also included the usual take-profit and stop-loss parameters in the event you want to play a little :)
We hope this helps you advance your personal system.
Happy Trading!
MA Streak Change ChannelChange Channel is like KC unless it uses percentage changes in price to set channel distance. Midline is zero-lag smoothed ROC with dynamic period based on MA Streak indicator, if MA Streak shows an ongoing trend, midline going strong and break out the channel.
Consider using ▲ green areas as a signal to buy and ▼ red areas as a sell signal. It works best in a flat market. Use in combination with other indicators.
Squeeze Momentum Signal Overlay [GN]Companion script for Squeeze Momentum Indicator that plots the signal on chart.
Relative Channel BandwidthThis indicator uses different volatility channels - Bollinger Band, Donchian Channel and Keltner Channel width to measure volatility.
Indicator plots channel bandwidth percentage with respect to close price.
This is not same as Bollinger Percent B - which is measure of where price is with respect to band. Instead this indicator is similar to ATR Percent indicator published here:
Plotting is color coded to indicate volatility zone:
Red : Extreme volatility
Orange : High volatility
Lime : Low volatility
Green : Extreme low volatility
These levels are again derived by long period bollinger bands
Adoptive Supertrend - BandsAnother adoption of supertrend. This time based on different channels - Bollinger Band, Keltner Channel, Donchian Channel and Pivot point based Donchian channel.
When price hits top of bands, it is considered as start or continuation of uptrend. When price hits bottom of the band it is considered as start or continuation of downtrend. Hence, supertrend is drawn based on these calculations. Use ATR Periods and ATR Multiplier to create stops certain ATR away from band's top and bottom.
Other supertrend adoptions published are here:
Pivot point based donchian channel is published here:
Multi Band ChannelPutting multiple bollinger bands/keltener channels together helps visualize the relative price movement. I have also used this in my BuyTheDip V2 strategy to measure dip and bounce back. 7 Standard deviation is used here. These are calculated based on the input fields StdDevStart and StdDevStep .
Screener - Mean Reversion Channel█ OVERVIEW
This is Screener script for Mean Reversion Channel Indicator
█ Description & How To Use
The screener works by scanning through up to 40 symbols and list down symbols that are currently within Overbought/Oversold Zone as defined by Mean Reversion Channel indicator.
The Overbought/Oversold Zone are further categorized and sorted by:
Strong : Indicated by "(Strong)" next to the symbol name
Normal : Indicated by the absence of "(Strong)" or "(Weak)" next to the symbol name
Weak : Indicated by "(Weak)" next to the symbol name
Notes: Refer to chart above to see how the Zone are categorized.
Notes: If the screener displays "Nothing Interesting". It simply means none of the screened assets are within the Overbought/Oversold Zone.
█ Features
- Scan up to 40 symbols at a time (By default, no asset is define. Once configured all the symbols you required, remember to save as default to save you from pain of configuring it again in the future)
- Options to scan by zones
- Custom Timeframe
█ Limitation
Due to multiple use of security() function required to call other symbols, expect the screener to be slow at certain times
█ Disclaimer
Past performance is not an indicator of future results.
My opinions and research are my own and do not constitute financial advice in any way whatsoever.
Nothing published by me constitutes an investment recommendation, nor should any data or Content published by me be relied upon for any investment/trading activities.
I strongly recommends that you perform your own independent research and/or speak with a qualified investment professional before making any financial decisions.
Any ideas to further improve this indicator are welcome :)
Credit: QuantNomad for his script idea on custom screener
Keltner Channel [LINKUSDT] 1HThis is a long-only strategy tested on LINK/USDT, 1 hour bar, from Feb 2019. The entry is determined by the breakout of upper Keltnel Channel and when the +DI is higher than 32. Instead of a fixed stop-loss from the original script , I change the exit to the middle band of the Keltnel Channel. 1st profit target will close 20% of the position. 2nd profit target will close 30% of the position. While the remaining 50% position will be closed when the price closes below the middle band of the Keltnel Channel, to take advantage of big trend. All parameters are adjustable. I added another option to enable or disable the ribbon trend filter.
My thoughts: For the same period, LINK appreciated 3000%. So I guess most in and out strategies couldn’t beat a buy and hold strategy during this period. But this doesn’t mean that this strategy is not feasible as each strategy is designed to only take advantage of a certain pattern or behavior of the market. Also, short term strategies allow you to use leverage and hence enable you to use you capital efficiently. Commission is set to 0.1%, taking account of the slippage.
Suggestion: Please perform walk forward analysis before you use real money for trading. Parameters need to be adjusted from time to time depends on your analysis. Can try using ATR for profit targets as over a longer term, the volatility might drop hence a high fixed % profit targets might not be realistic.
Any suggestions are welcome!
Mean Reversion Channel - (fareid's MRI Variant)Description :
Mean Reversion Channel objective, based on Mean Reversion theory ( everything has a tendency to revert back to its mean), is to help visualizing:
Inner Channel -> Dynamic Support and Resistance
Outer Channel -> Overbought/Oversold Zone which may signal consolidation phase or potential reversal due to unsustainable move
Details on some of the filtering type used for mean calculation can be read in Ehlers Technical Papers: "Swiss Army Knife Indicator" and/or his book "Cybernetics Analysis for Stock and Futures"
Disclaimer:
These study scripts was built only to test/visualize an idea to see its viability and if it can be used to optimize existing strategy.
Any ideas to further improve this indicator are welcome :)
Spread Entry StrengthThis is an overlay indicator showing a strong potential for entry into an option spread trade.
2 background shadings will occur:
The background will shade blue if the ticker is prime for a Bullish Call spread.
The background will shade purple if the the ticker is prime for a Bearish Put spread.
In theory, if the SE Strength is at one of the extremes of the Bear or Bull side, then a spread is prime for entry.
To calculate this, 8 conditions receive a 1 or zero dependent on whether the condition is true (1) or false (0), and then all of those are summed. The primary gist of the strength comes from Nishant's book, or my interpretation thereof, with some additives that limits what I need to review (such as condition 8 below.)
The 8 Bull Conditions are:
1) Bollinger Bands are outside of the Keltner Channels
2) ADX is trending up
3) RSI is trending up
4) -DI is trending down
5) RSI is under 30
6) Price is below the lower Keltner Channel
7) Price is between the lower Bollinger Band and the Bollinger basis.
8) Price at one point within the last 5 bars was below the lower Bollinger Band
The 8 Bear Conditions are the inverse conditions (except the first):
1) Bollinger Bands are outside of the Keltner Channels
2) ADX is trending down
3) RSI is trending down
4) +DI is trending up
5) RSI is over 70
6) Price is above the upper Keltner Channel
7) Price is between the upper Bollinger Band and the Bollinger basis.
8) Price at one point within the last 5 bars was above the upper Bollinger Band
There is a "market noise" filter that will filter out shading when another market move is considered, i.e. if you don't want to see the potential trade when QQQ moves more than 1% then do the following in the settings:
Check "Market Filter"
Enter QQQ in the "Market Ticker To Use"
Enter 1 in the "Market Too Hot Level"
Press Ok
Obviously, the same holds true for the "Market Too Cool Filter."
Spread Entry Balance of PowerThis is a bar chart showing the strength of a potential option spread entry using 8 conditions for each side of a trade, bull or bear.
In theory, if the SE Strength (Spread Entry Strength) is at one of the extremes of the Bear or Bull side, then a spread is prime for entry.
To calculate this, the 8 conditions receive a 1 or zero dependent on whether the condition is true (1) or false (0), and then all of those are summed. The primary gist of the strength comes from Nishant's book, or my interpretation thereof, with some additives that limits what I need to review (such as condition 8 below.)
The 8 Bull Conditions are:
1) Bollinger Bands are outside of the Keltner Channels
2) ADX is trending up
3) RSI is trending up
4) -DI is trending down
5) RSI is under 30
6) Price is below the lower Keltner Channel
7) Price is between the lower Bollinger Band and the Bollinger basis.
8) Price at one point within the last 5 bars was below the lower Bollinger Band
The 8 Bear Conditions are the inverse conditions (except the first), and the conditions are given a negative disposition (meaning they sum to -8 :)):
1) Bollinger Bands are outside of the Keltner Channels
2) ADX is trending down
3) RSI is trending down
4) +DI is trending up
5) RSI is over 70
6) Price is above the upper Keltner Channel
7) Price is between the upper Bollinger Band and the Bollinger basis.
8) Price at one point within the last 5 bars was above the upper Bollinger Band
Square Root Moving AverageAbstract
This script computes moving averages which the weighting of the recent quarter takes up about a half weight.
This script also provides their upper bands and lower bands.
You can apply moving average or band strategies with this script.
Introduction
Moving average is a popular indicator which can eliminate market noise and observe trend.
There are several moving average related strategies used by many traders.
The first one is trade when the price is far from moving average.
To measure if the price is far from moving average, traders may need a lower band and an upper band.
Bollinger bands use standard derivation and Keltner channels use average true range.
In up trend, moving average and lower band can be support.
In ranging market, lower band can be support and upper band can be resistance.
In down trend, moving average and upper band can be resistance.
An another group of moving average strategy is comparing short term moving average and long term moving average.
Moving average cross, Awesome oscillators and MACD belong to this group.
The period and weightings of moving averages are also topics.
Period, as known as length, means how many days are computed by moving averages.
Weighting means how much weight the price of a day takes up in moving averages.
For simple moving averages, the weightings of each day are equal.
For most of non-simple moving averages, the weightings of more recent days are higher than the weightings of less recent days.
Many trading courses say the concept of trading strategies is more important than the settings of moving averages.
However, we can observe some characteristics of price movement to design the weightings of moving averages and make them more meaningful.
In this research, we use the observation that when there are no significant events, when the time frame becomes 4 times, the average true range becomes about 2 times.
For example, the average true range in 4-hour chart is about 2 times of the average true range in 1-hour chart; the average true range in 1-hour chart is about 2 times of the average true range in 15-minute chart.
Therefore, the goal of design is making the weighting of the most recent quarter is close to the weighting of the rest recent three quarters.
For example, for the 24-day moving average, the weighting of the most recent 6 days is close to the weighting of the rest 18 days.
Computing the weighting
The formula of moving average is
sum ( price of day n * weighting of day n ) / sum ( weighting of day n )
Day 1 is the most recent day and day k+1 is the day before day k.
For more convenient explanation, we don't expect sum ( weighting of day n ) is equal to 1.
To make the weighting of the most recent quarter is close to the weighting of the rest recent three quarters, we have
sum ( weighting of day 4n ) = 2 * sum ( weighting of day n )
If when weighting of day 1 is 1, we have
sum ( weighting of day n ) = sqrt ( n )
weighting of day n = sqrt ( n ) - sqrt ( n-1 )
weighting of day 2 ≒ 1.414 - 1.000 = 0.414
weighting of day 3 ≒ 1.732 - 1.414 = 0.318
weighting of day 4 ≒ 2.000 - 1.732 = 0.268
If we follow this formula, the weighting of day 1 is too strong and the moving average may be not stable.
To reduce the weighting of day 1 and keep the spirit of the formula, we can add a parameter (we call it as x_1w2b).
The formula becomes
weighting of day n = sqrt ( n+x_1w2b ) - sqrt ( n-1+x_1w2b )
if x_1w2b is 0.25, then we have
weighting of day 1 = sqrt(1.25) - sqrt(0.25) ≒ 1.1 - 0.5 = 0.6
weighting of day 2 = sqrt(2.25) - sqrt(1.25) ≒ 1.5 - 1.1 = 0.4
weighting of day 3 = sqrt(3.25) - sqrt(2.25) ≒ 1.8 - 1.5 = 0.3
weighting of day 4 = sqrt(4.25) - sqrt(3.25) ≒ 2.06 - 1.8 = 0.26
weighting of day 5 = sqrt(5.25) - sqrt(4.25) ≒ 2.3 - 2.06 = 0.24
weighting of day 6 = sqrt(6.25) - sqrt(5.25) ≒ 2.5 - 2.3 = 0.2
weighting of day 7 = sqrt(7.25) - sqrt(6.25) ≒ 2.7 - 2.5 = 0.2
What you see and can adjust in this script
This script plots three moving averages described above.
The short term one is default magenta, 6 days and 1 atr.
The middle term one is default yellow, 24 days and 2 atr.
The long term one is default green, 96 days and 4 atr.
I arrange the short term 6 days to make it close to sma(5).
The other twos are arranged according to 4x length and 2x atr.
There are 9 curves plotted by this script. I made the lower bands and the upper bands less clear than moving averages so it is less possible misrecognizing lower or upper bands as moving averages.
x_src : how to compute the reference price of a day, using 1 to 4 of open, high, low and close.
len : how many days are computed by moving averages
atr : how many days are computed by average true range
multi : the distance from the moving average to the lower band and the distance from the moving average to the lower band are equal to multi * average true range.
x_1w2b : adjust this number to avoid the weighting of day 1 from being too strong.
Conclusion
There are moving averages which the weighting of the most recent quarter is close to the weighting of the rest recent three quarters.
We can apply strategies based on moving averages. Like most of indicators, oversold does not always means it is an opportunity to buy.
If the short term lower band is close to the middle term moving average or the middle term lower band is close to the long term moving average, it may be potential support value.
References
Computing FIR Filters Using Arrays
How to trade with moving averages : the eight trading signals concluded by Granville
How to trade with Bollinger bands
How to trade with double Bollinger bands
