Standard Error Bands (SEB)

Prev Next


The Standard Error Bands (SEB) indicator shows the direction of the current trend by plotting a linear regression average value (middle line) and the volatility deviation by plotting upper and lower bands. The deviation from the linear regression indicates increased volatility and risk. The indicator is similar to Bollinger Bands (BB) with the only difference that SEB uses smoothed linear regression instead of moving average that is used in BB.

The middle line is calculated as the linear regression for each bar:

y = a + bx
b = (nΣxy - ΣxΣy) / ( nΣx2 - (Σx)2 ))
a = (Σy - bΣx) / n

where:
y – the data price source
x – the number of bars
a – the constant (the value when x equals zero)
b – the slope of the line
n – the number of data points selected

The standard error bands are calculated as follows:

UpperBand = MidLine + 2 x StdError
LowerBand = MidLine - 2 x StdError

where:
StdError – the aproximate standard deviation

Standard Error BandsStandard Error Bands

INPUTS

Price
The following data sources are available for the price:
  • Open. The Open Price of every bar for the specified period.
  • High. The High Price of every bar for the specified period. 
  • Low. The Low Price of every bar for the specified period.
  • Close. The Close Price of every bar for the specified period. 
  • Median. The Median Price of every bar for the specified period - (High + Low) / 2.
  • Typical. The sum of High, Low, and Close prices divided by 3 for the specified period- (High + Low + Close) / 3.
  • OHLC average. The arithmetical mean of High, Low, Open, and Close prices for the specified period - (Open + High + Low + Close) / 4.
  • Volume. The Close Price of every bar + average volume of an asset for the specified period.
InputDescription
DisplaceThe displacement of the indicator in bars
LinRegLengthThe number of bars used to calculate the Linear Regression (MidLine)
SmLengthThe number of bars used to smooth the Linear Regression (MidLine)
NumDevUpThe multiplier for the stardard error band used in the upward shift calculation 
NumDevDownThe multiplier for the standard error band used in the downward shift calculation

PLOTS

The plot renders the data you are working with on the chart. You can show/hide a plot by clicking the corresponding item in the settings. Every plot has a set of basic settings that you can change: color, weight, and type.

PlotDescription
LowerBandThe lower standard error band
MidLineThe Linear Regression plot
UpperBandThe upper stardard error band 
Color

Click the color rectangle under the plot's name to open the palette. Use the slider at the bottom to set the opacity of the color.

Palette

To create a custom color:

  1. Clickto open the color picker
  2. By using the slider at the bottom, adjust the main color
  3. Then, set a shade for the main color on the palette

The custom-created colors are added to your palette. To remove a custom color, drag it out of the palette.

Weight
Change the value (in px) to adjust the thickness of the plot.
Type
The following plot types are available:
  • Linear. The plot is displayed as a line.
  • Column. The plot is displayed as a histogram.
  • Difference. The plot is displayed as a line which represents the difference between High and Low prices of the bar.
  • Points. The plot is displayed as points.
  • Text above candle. The plot is displayed as the price labels above candles on the chart.
  • Text below candle. The plot is displayed as the price labels below candles on the chart.
  • Triangle above candle. The plot is displayed as triangles above candles on the chart.
  • Color candle. The candles on candle-type charts are highlighted with specified colors.

OVERLAYING

Check Overlaying to display the indicator on the chart. Otherwise, the indicator is shown in a study pane down below.

Note: To reset the settings, click Restore to Default next to the SETTINGS: INDICATOR'S TITLE