Candlestick

Candlestick is a type of chart commonly used in financial markets to represent the price movement of an asset, such as stocks or cryptocurrencies. It consists of individual "candles" that display the opening, closing, high, and low prices for a specific time period. Each candle has a rectangular body, representing the opening and closing prices, and thin lines, called wicks or shadows, indicating the highest and lowest prices during that time frame.

Candlestick charts often use colors to visually indicate whether the value within a specific candle has increased or decreased. For example, a positive price movement is typically represented by a green candle, indicating that the closing price is higher than the opening price (and by a red color otherwise). Basically "candlestick" main statistic is an indicator of whether the price has increased.

This guide describes comprehensively the process of building candlestick chart and all the details of its customization.

This notebook uses definitions from DataFrame.

Usage

Binning is commonly used in statistics and data analysis to simplify complex data sets and make them easier to interpret. Histogram (or any other plot with "bin" statistics) helps to give an overview of the sample distribution.

Arguments

Input (mandatory):
- x — candle x-position (often time or date describing)
- open — candle open value
- high — candle high value
- low — candle low value
- close — candle close value

Generalized signature

The specific signature depends on the function, but all functions related to "candlestick" statistic (which will be discussed further below - different variations of statCandlestick(), candlestick()) have approximately the same signature with the arguments above:

statCandlestickArgs := 
   x, open, high, low, close

The possible types of x, open, high, low, close depend on where a certain function is used. They can be simply Iterable (List, Set, etc.) or a reference to a column in a DataFrame (String, ColumnAccessor) or the DataColumn itself.

Output statistics

name	type	description
Stat.x	X	Candle `x`-position
Stat.open	Double	Candle open. Equals to input `open`.
Stat.close	Double	Candle close. Equals to input `close`.
Stat.min	Double	Candle minimum. Equals to input `low`.
Stat.lower	Double	Candle lower value (i.e. smaller of `open`/`close`)
Stat.upper	Double	Candle lower value (i.e. greater of `open`/`close`)
Stat.max	Double	Candle maximum. Equals to input `high`.
Stat.isIncreased	Boolean	Increase indicator: `true` if `close` value is greater than `open` value

StatCandlestick plots

// Create a simple candlestick dataset
val xList = listOf("Jan", "Feb", "Mar", "Apr", "May")
val openList = listOf(14.2, 6.7, 8.8, 11.2, 4.0)
val highList = listOf(15.5, 9.6, 10.7, 11.7, 9.9)
val lowList = listOf(7.5, 6.1, 8.5, 5.4, 4.0)
val closeList = listOf(8.0, 8.6, 10.7, 6.5, 9.8)
// Gather lists into df as columns
val df = dataFrameOf(
    "x" to xList,
    "open" to openList,
    "high" to highList,
    "low" to lowList,
    "close" to closeList,
)
df.head()

x	open	high	low	close
Jan	14.2	15.5	7.5	8.0
Feb	6.7	9.6	6.1	8.6
Mar	8.8	10.7	8.5	10.7
Apr	11.2	11.7	5.4	6.5
May	4.0	9.9	4.0	9.8

df has a signature

x	open	high	low	close

Let's take a look at StatCandlestick output DataFrame:

df.statCandlestick("x", "open", "high", "low", "close")

Stat

x	open	close	min	lower	upper	max	isIncreased
Jan	14.2	8.0	7.5	8.0	14.2	15.5	false
Feb	6.7	8.6	6.1	6.7	8.6	9.6	true
Mar	8.8	10.7	8.5	8.8	10.7	10.7	true
Apr	11.2	6.5	5.4	6.5	11.2	11.7	false
May	4.0	9.8	4.0	4.0	9.0	9.9	true

It has the following signature:

Stat
x	open	close	min	lower	upper	max	isIncreased

As you can see, we got a DataFrame with one ColumnGroup called Stat which contains several columns with statics. For statCandlestick, each row corresponds to one candle. Stat.x is the candle x-coordinate. Stat.open and Stat.close correspond to candle open and close. Stat.min and Stat.max correspond to candle low and high. Stat.lower and Stat.upper correspond to candle edges. Stat.isIncreased shows if value is increased (i.e close > open).

DataFrame with "candlestick" statistics is called StatCandlestickFrame

`statCandlestick` context transform

statCandlestick(statCandlestickArgs) { /*new plotting context*/ } modifies a plotting context — instead of original data (no matter was it empty or not) new StatCandlestick dataset (calculated on given arguments. Inputs and weights can be provided as Iterable or as dataset column reference — by name as a String, as a ColumnReference or as a DataColumn) is used inside a new context (original dataset and primary context are not affected — you can add layers using initial dataset outside the statCandlestick context). Since the old dataset is irrelevant, we cannot use references for its columns. But we can refer to the new ones. They are all contained in the Stat group and can be called inside the new context:

plot {
    statCandlestick(xList, openList, lowList, highList, closeList) {
        errorBars {
            x(Stat.x)
            yMin(Stat.lower)
            yMax(Stat.upper)
            borderLine.color(Stat.isIncreased) {
                scale = categorical(true to Color.GREEN, false to Color.RED)
                legend.type = LegendType.None
            }
        }
    }
}

Candlestick layer

Basically, candlestick plot is a box plot where each box represents one candle. Box whisker's ends correspond to low and high values; and lower and upper edges to open and close (so here we need to determine which is greater and which is lesser — that's what we counted in the statistics). Non-positional attributes (most often color) indicate whether an increase or decrease has occurred. So basically, we can build a candlestick with statCandlestick and boxes as follows:

val statCandlestickBoxesPlot = df.plot {
    statCandlestick("x", "open", "high", "low", "close") {
        boxes {
            x(Stat.x)
            yMin(Stat.min)
            lower(Stat.lower)
            upper(Stat.upper)
            yMax(Stat.max)
            // temporary solution, middle shoudn't be necceasary
            middle(List(Stat.x.size()) { null })
            val colorScale = Scale.categorical(true to Color.GREEN, false to Color.RED)
            fillColor(Stat.isIncreased) {
                scale = colorScale
                // remove legend
                legend.type = LegendType.None
            }
            borderLine.color(Stat.isIncreased) {
                scale = colorScale
                // remove legend
                legend.type = LegendType.None
            }
            alpha = 0.6
            // remove whisker ends
            whiskerWidth = 0.0
        }
    }
    layout.title = "`statCandlestick` + `boxes`"
}
statCandlestickBoxesPlot

Candlestick Plot using StatCandlestick with Boxes

But we can do it even faster with candlestick(statCandlestickArgs) method:

val candlestickPlot = plot {
    candlestick(xList, openList, highList, lowList, closeList)
    layout.title = "`candlestick`"
}
candlestickPlot

Candlestick Plot using Candlestick Layer

Let's compare them:

plotGrid(listOf(statCandlestickBoxesPlot, candlestickPlot))

These two plots are almost identical (the only difference in tooltips). Indeed, candlestick just uses statCandlestick and boxes and performs aesthetic mappings under the hood.

`candlestick` customization

We can customize candlestick layer: candlestick() optionally opens a new context, where we can configure it. We can set different color, borderline color, etc. for candles with increasing and decreasing value with special DSL, or make general settings (as in the usual context opened by boxes { ... }):

df.plot {
    candlestick(x, open, high, low, close) {
        // Boxes context + StatCandlestick context
        // change fill color when increased
        increase {
            fillColor = Color.hex("#00FFFF")
        }
        // change fill color when decreased
        decrease.fillColor = Color.hex("#FF0000")
        // set constant border line color for all candles
        borderLine.color = Color.GREY
    }
}

However, it can also be done as with other statistical layers API (i.e., through mappings from statistics from StatCandlestick dataset):

df.plot {
    candlestick(x, open, high, low, close) {
        // Boxes context + StatCandlestick context
        alpha(Stat.isIncreased) {
            scale = categorical(true to 1.0, false to 0.1)
            legend {
                name = ""
                breaksLabeled(true to "increase", false to "decrease")
            }
        }
        fillColor = Color.GREY
        borderLine.color = Color.GREY
    }
}

`candlestick` plot

candlestick(statCandlestickArgs) and DataFrame.candlestick(statCandlestickArgs) are a family of functions for fast plotting a candlestick.

candlestick(xList, openList, highList, lowList, closeList)

df.candlestick("x", "open", "high", "low", "close")

In case you want to provide inputs using column selection DSL, it’s a bit different from the usual one — you should assign them throw invocation eponymous functions:

df.candlestick {
    x(x)
    open(open)
    high(high)
    low(low)
    close(close)
}

Candlestick plot can be configured with .configure {} extension — it opens a context that combines bars, StatCandlestick and plot context. That means you can configure boxes settings, mappings using StatCandlestick dataset and any plot adjustments:

df.candlestick("x", "open", "high", "low", "close").configure {
    // Boxes + StatCandlestick + PlotBuilder
    // Can't add a new layer
    y.limits = 3.0..17.0
    increase {
        borderLine.color = Color.BLUE
    }
    decrease.borderLine.color = Color.YELLOW
    borderLine.width = 2.5
    fillColor = Color.GREY
    alpha = 0.6
    // Can configure general plot adjustments
    layout {
        title = "Configured candlestick plot"
        size = 800 to 400
    }
}

Last modified: 15 July 2024