# RedK Volume-Accelerated Directional Energy Ratio

Author: 张超, Date: 2022-05-18 15:21:57
Tags: EMA sma WMA

The Volume-Accelerated Directional Energy Ratio (VADER) makes use of price moves (displacement) and the associated volume (effort) to estimate the positive (buying) and negative (selling) “energy” behind the scenes, enabling traders to “read the market action” in more details and adjust their trading decisions accordingly.

I have always been a fan of technical analysis concepts that are simple, and that integrate both price action and volume together - The concept behind VADER is really a simple one.

Let’s walk though it as we avoid getting too technical: Large price moves that are associated with large volume means buyers (if the move is up) or sellers (when the move is down) are serious and are “in control” of the action On the other hand, when the price moves are small but with large volume , it means there’s a fight, or more of a balance of energy, between buying and selling. Also when large price moves are associated with relatively limited volume , there’s a lack of “energy” from either buyers or sellers - and moves likes these are usually short-lived.

The analogy with VADER, is that we look at price moves (change of close between 2 bars) as the displacement (or action result) and the associated volume as the “effort” behind this action – Combining these 2 values together, the displacement and the effort, gives us a representation or a proxy of the underlying energy (in a specific direction). when both values (displacement and effort) are high, then the resulting energy is high - and if one of these values are low, the resulting energy is low.

we then take an average of that relative energy in each direction (positive = buying and negative = selling) and calculate the net energy.

note that we’re approaching the analogy here from a trading perspective and not from physics perspective :) – we can be forgiven if the energy calculation in physics is different …

backtest

```/*backtest
start: 2022-04-17 00:00:00
end: 2022-05-16 23:59:00
period: 1h
basePeriod: 15m
exchanges: [{"eid":"Futures_Binance","currency":"BTC_USDT"}]
*/

// This source code is subject to the terms of the Mozilla Public License 2.0 at https://mozilla.org/MPL/2.0/

//@version=5
indicator('RedK Volume-Accelerated Directional Energy Ratio', 'RedK VADER v3.0', precision=0, timeframe='', timeframe_gaps=false)

// ***********************************************************************************************************
// Choose volume calculation method.. Relative vs full.
// Relative magnifies effect of recent volume spikes (up or down)
f_RelVol(_value, _length) =>
min_value = ta.lowest(_value, _length)
max_value = ta.highest(_value, _length)
ta.stoch(_value, max_value, min_value, _length) / 100
// ***********************************************************************************************************

// ***********************************************************************************************************
// Choose MA type for the base DER calculation ..
// WMA is my preference and is default .. SMA is really slow and lags a lot - but added for comparison
f_derma(_data, _len, MAOption) =>
value =
MAOption == 'SMA' ? ta.sma(_data, _len) :
MAOption == 'EMA' ? ta.ema(_data, _len) :
ta.wma(_data, _len)
// ***********************************************************************************************************

// ===========================================================================================================
//      Inputs
// ===========================================================================================================

price   = close
length  = input.int(9, minval=1)
DER_avg = input.int(5, 'Average', minval=1, inline='DER', group='Directional Energy Ratio')
MA_Type = input.string('WMA', 'DER MA type', options=['WMA', 'EMA', 'SMA'], inline='DER', group='Directional Energy Ratio')
smooth  = input.int(3, 'Smooth', minval=1,  inline='DER_1', group='Directional Energy Ratio')

show_senti = input.bool(false, 'Sentiment',  inline='DER_s', group='Directional Energy Ratio')
senti   = input.int(20, 'Length', minval=1, inline='DER_s', group='Directional Energy Ratio')

v_calc  = input.string('Relative', 'Calculation', options=['Relative', 'Full', 'None'], group='Volume Parameters')
vlookbk = input.int(10, 'Lookback (for Relative)', minval=1,                            group='Volume Parameters')

// ===========================================================================================================
//          Calculations
// ===========================================================================================================

// Volume Calculation Option  -- will revert to no volume acceleration for instruments with no volume data
vola    =
v_calc == 'None' or na(volume) ? 1 :
v_calc == 'Relative' ? f_RelVol(volume, vlookbk) :
volume

R       = (ta.highest(2) - ta.lowest(2)) / 2                    // R is the 2-bar average bar range - this method accomodates bar gaps
sr      = ta.change(price) / R                                  // calc ratio of change to R
rsr     = math.max(math.min(sr, 1), -1)                         // ensure ratio is restricted to +1/-1 in case of big moves
c       = fixnan(rsr * vola)                                    // add volume accel -- fixnan adresses cases where no price change between bars

c_plus  = math.max(c, 0)                                        // calc directional vol-accel energy
c_minus = -math.min(c, 0)

// plot(c_plus)
// plot(c_minus)

avg_vola    = f_derma(vola, length, MA_Type)
dem         = f_derma(c_plus, length, MA_Type)  / avg_vola          // directional energy ratio
sup         = f_derma(c_minus, length, MA_Type) / avg_vola

adp         = 100 * ta.wma(dem, DER_avg)                            // average DER
asp         = 100 * ta.wma(sup, DER_avg)
anp         = adp - asp                                             // net DER..
anp_s       = ta.wma(anp, smooth)

// Calculate Sentiment - a VADER for a longer period and can act as a baseline (compared to a static 0 value)
// note we're not re-calculating vol_avg, demand or supply energy for sentiment. this would've been a different approach
s_adp       = 100 * ta.wma(dem, senti)                            // average DER for sentiment length
s_asp       = 100 * ta.wma(sup, senti)
V_senti     = ta.wma(s_adp - s_asp, smooth)

// ===========================================================================================================
//      Colors & plots
// ===========================================================================================================
c_asp   = color.new(color.orange, 30)
c_fd    = color.new(color.green, 80)
c_fs    = color.new(color.red, 80)
c_zero  = color.new(#ffee00, 70)

c_up    = color.new(#359bfc, 0)
c_dn    = color.new(#f57f17, 0)

c_sup   = color.new(#33ff00, 80)
c_sdn   = color.new(#ff1111, 80)
up      = anp_s >= 0
s_up    = V_senti >=0

hline(0, 'Zero Line', c_zero, hline.style_solid)

// =============================================================================
// v3.0 --- Sentiment will be represented as a 4-color histogram
c_grow_above = #1b5e2080
c_grow_below = #dc4c4a80
c_fall_above = #66bb6a80
c_fall_below = #ef8e9880

sflag_up = math.abs(V_senti) >= math.abs(V_senti[1])

plot(show_senti ? V_senti : na, "Sentiment", style=plot.style_columns,
color = s_up ? (sflag_up ? c_grow_above : c_fall_above) :
sflag_up ? c_grow_below : c_fall_below)
// =============================================================================

s = plot(asp, 'Supply Energy', c_asp, 2, style=plot.style_circles,  join=true)
fill(d, s, adp > asp ? c_fd : c_fs)

plot(anp_s, 'Signal', up ? c_up : c_dn, 3)

// ===========================================================================================================
// ===========================================================================================================

// "." in alert title for the alerts to show in the right order up/down/swing

// ===========================================================================================================
// ===========================================================================================================

v_speedup = ta.crossover(anp_s, V_senti)
v_slowdn  = ta.crossunder(anp_s, V_senti)