Usage

Basic commands and recommended usage for custom and bulk processing.

General#

Version#

Prints out the version.

1
2

import pandas_ta as ta
print(ta.version)

Help#

Ways to utilize Python's help() command for the library.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11

import pandas as pd
import pandas_ta as ta

# Create an empty DataFrame to access the "ta" extension.
df = pd.DataFrame()

# Indicator Help
help(ta.macd)

# Help about the DataFrame Extension "ta"
help(df.ta)

Indicators and Library Version#

Prints out the list of indicators.

1
2
3
4
5
6
7

import pandas as pd
import pandas_ta as ta

# Create an empty DataFrame to access the "ta" extension.
df = pd.DataFrame()

print(df.ta.indicators())

Programming Conventions#

This library supports three programming conventions: Standard "TA Lib" Convention, Pandas "ta" DataFrame Extension Convention and the Pandas "ta" study() Convention.

Each "Programming Convention" showcases:

• The use of default and user specified parameters.

• Enabling/disabling ta-lib.

  Standard OHLCV Source Indicators Examples

  • ema()
    • Default parameters
  • macd(fast=5, slow=20, talib=False)
    • Disable TA Lib computation

• Indicator chaining/composition.

  Chained/Composed Indicators Examples

  • ohlc4()
    • Source
  • ema(close=ohlc4)
    • Use ohlc4 as input instead of close
  • macd(close=ohlc4, fast=5, slow=20, talib=False)
    • Disable TA Lib computation

Example Setup Code#

Also examples assumes yfinance returns at least the following column names: "Open", "High", "Low", "Close", and "Volume"; additional columns are not needed for the library.

1
2
3
4
5
6
7
8

import pandas as pd
import pandas_ta as ta

_df = pd.DataFrame()

df = # df with ohlcv columns
# OR
df = _df.ta.ticker("spy", period="5y")

Standard Convention#

With the above "Example Setup Code", Standard Convention is similar to TA Lib but with lowercase indicator names. The Standard Convention requires more explicit handling of indicator inputs and outputs as well as concatenating to the DataFrame (if needed).

1
2
3
4
5
6

# Since ema() returns a Series, it can immediately be concatenated to the df
df["EMA_10"] = ta.ema(df["Close"])

# Since macd() returns a DataFrame, it must be manually concatenated to the df
macddf = ta.macd(df["Close"], fast=5, slow=20, talib=False)
df = pd.concat([df, macddf])

 1
 2
 3
 4
 5
 6
 7
 8
 9
10

ohlc4 = ohlc4(df["Open"], df["High"], df["Low"], df["Close"])

# Since ema() returns a Series, it can immediately be concatenated to the df
df["EMAohlc4_10"] = ema(close=ohlc4)

# Since macd() returns a DataFrame, it must be manually concatenated to the df
macddf = macd(close=ohlc4, fast=5, slow=20, talib=False)
# Alternatively ...
# macddf = macd(close=df["EMAohlc4_10"], fast=5, slow=20, talib=False)
df = pd.concat([df, macddf])

DataFrame "ta" Extension Convention#

With the above "Example Setup Code", the DataFrame "ta" Extension Convention simplifies the number of steps to calculate and concatenate the indicators onto the df.

1
2

df.ta.ema(append=True)
df.ta.macd(fast=5, slow=20, talib=False, append=True)

1
2
3
4
5
6
7
8
9

# One way ...
df.ta.ohlc4(append=True)
df.ta.ema(close=df["OHLC4"], append=True)
df.ta.macd(close=df["OHLC4"], fast=5, slow=20, talib=False, append=True)

# Alternatively ...
ohlc4 = df.ta.ohlc4(append=True)
df.ta.ema(close=ohlc4, append=True)
df.ta.macd(close=ohlc4, fast=5, slow=20, talib=False, append=True)

Pandas "ta" Study Convention#

With the above "Example Setup Code", the Pandas "ta" study() Convention also assumes the following columns: "Open", "High", "Low", "Close", and "Volume" exist in the df.

1
2
3
4
5
6
7
8
9

std_example = ta.Study(
    name="Standard OHLCV Example",
    cores=0,
    ta=[
        {"kind": "ema"},
        {"kind": "macd", "fast": 5, "slow": : 20, "talib": False}
    ]
)
df.ta.study(std_example, verbose=True)

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14

chained_example = ta.Study(
    name="Chained Example",
    cores=0,
    ta=[
        { "kind": "ohlc4" },
        { "kind": "ema", "close": "OHLC4", "prefix": "OHLC4" },
        {
            "kind": "macd", "close": "OHLC4",
            "fast": 5, "slow": 20,
            "talib": False, "prefix": "OHLC4"
        }
    ]
)
df.ta.study(chained_example, verbose=True)