Candlestick charts

Candlestick charts are commonly used in the stock market business. For a given period, the open and close prices of some stock are represented as a vertical bar, and vertical line emerging from that box shows the maximum and minimum values of that stock for the period.

Taipy creates a candlestick chart when the control's type property is set to "candlestick".

Key properties

Name	Value	Notes
type	candlestick
x	x values
open	Open values
close	Close values
low	Low values
high	High values
options	dictionary	decreasing and increasing can be used to customize the rendering of candlesticks.
*layout	dictionary	xaxis can be used to define how what the x axis should display and how.

Examples

Simple candlestick chart

You can download the entire source code used in this section from the GitHub repository.

The following example creates a candlestick chart for the AAPL stock price. It shows all the relevant daily information for an entire month.

Note that we use the Python yfinance package to query the historical data:

# Extraction of a month of stock data for AAPL using the
# yfinance package (see https://pypi.org/project/yfinance/).
ticker = yfinance.Ticker("AAPL")
# The returned value is a Pandas DataFrame.
stock = ticker.history(interval="1d", start="2018-08-01", end="2018-08-31")
# Copy the DataFrame's index to a new column
stock["Date"] = stock.index

The x axis will be appropriately labeled "Date", thanks to the additional column.

The chart definition is the following:

Definition

MarkdownHTMLPython

<|{stock}|chart|type=candlestick|x=Date|open=Open|close=Close|low=Low|high=High|>

<taipy:chart type="candlestick" x="Date" open="Open" close="Close" low="Low" high="High">{stock}</taipy:chart>

import taipy.gui.builder as tgb
...
tgb.chart("{stock}", type="candlestick", x="Date", open="Open", close="Close", low="Low", high="High")

The chart is displayed like this:

Simple candlestick chart

Styling candlestick charts

You can download the entire source code used in this section from the GitHub repository.

By default, Plotly creates green candlestick bars when the value grows (when close is greater than open) and red candlestick bars when the value decreases.

Here is some code that applies colors of our choosing to the candlesticks. This code also removes the range slider (the thin candlestick overview under the main area):

# Extraction of a few days of stock historical data for AAPL using
# the yfinance package (see https://pypi.org/project/yfinance/).
# The returned value is a Pandas DataFrame.
ticker = yfinance.Ticker("AAPL")
stock = ticker.history(interval="1d", start="2018-08-18", end="2018-09-10")
# Copy the DataFrame index to a new column
stock["Date"] = stock.index

options = {
    # Candlesticks that show decreasing values are orange
    "decreasing": {
        "line": {
            "color": "orange"
        }
    },
    # Candlesticks that show decreasing values are blue
    "increasing": {
        "line": {
            "color": "blue"
        }
    }
}

layout = {
    "xaxis": {
        # Hide the range slider
        "rangeslider": {
            "visible": False
        }
    }
}

The data set is similar yet smaller than in the first example.

We have added two dictionaries (options and layout) that let us customize how the chart is ultimately rendered.

We use these dictionaries in the chart definition:

Definition

MarkdownHTMLPython

<|{stock}|chart|type=candlestick|x=Date|open=Open|close=Close|low=Low|high=High|options={options}|layout={layout}|>

<taipy:chart type="candlestick" x="Date" open="Open" close="Close" low="Low" high="High" options="{options}" layout="{layout}">{stock}</taipy:chart>

import taipy.gui.builder as tgb
...
tgb.chart("{stock}", type="candlestick", x="Date", open="Open", close="Close", low="Low", high="High", options="{options}", layout="{layout}")

This is what the customized candlestick chart looks like:

Styling candlestick charts

Using a time series

You can download the entire source code used in this section from the GitHub repository.

So far, the x values were taken from the Pandas DataFrame that was built by the yfinance API: the index of the DataFrame contains a series of Timestamp instances from the Pandas library.

You can create your own time series using the common datetime type that are available in all Python implementations in the datetime package:

# Retrieved history:
# (Open, Close, Low, High)
stock_history = [
    (311.05, 311.00, 310.75, 311.33),
    (308.53, 308.31, 307.72, 309.00),
    (307.35, 306.24, 306.12, 307.46),
    ...
    (299.42, 300.50, 299.42, 300.50),
    (300.70, 300.65, 300.32, 300.75),
    (300.65, 299.91, 299.91, 300.76)
]
start_date = datetime.datetime(year=2022, month=10, day=21)
period = datetime.timedelta(seconds=4 * 60 * 60)  # 4 hours

data = {
    # Compute date series
    "Date": [start_date+n*period for n in range(0, len(stock_history))],
    # Extract open values
    "Open": [v[0] for v in stock_history],
    # Extract close values
    "Close": [v[1] for v in stock_history],
    # Extract low values
    "Low": [v[2] for v in stock_history],
    # Extract high values
    "High": [v[3] for v in stock_history]
}

The chart definition is straightforward:

Definition

MarkdownHTMLPython

<|{stock}|chart|type=candlestick|x=Date|open=Open|close=Close|low=Low|high=High|>

<taipy:chart type="candlestick" x="Date" open="Open" close="Close" low="Low" high="High">{stock}</taipy:chart>

import taipy.gui.builder as tgb
...
tgb.chart("{stock}", type="candlestick", x="Date", open="Open", close="Close", low="Low", high="High")

As you can see, Taipy deals with datetime objects as well as specific types from Pandas, transparently.

Here is the result:

Using a time series