# -*- coding: utf-8 -*-
from numpy import nan as npNaN
from pandas import DataFrame
from pandas_ta.momentum import mom
from pandas_ta.overlap import ema, linreg, sma
from pandas_ta.trend import decreasing, increasing
from pandas_ta.volatility import bbands, kc
from pandas_ta.utils import get_offset
from pandas_ta.utils import unsigned_differences, verify_series
[docs]def squeeze(high, low, close, bb_length=None, bb_std=None, kc_length=None, kc_scalar=None, mom_length=None, mom_smooth=None, use_tr=None, mamode=None, offset=None, **kwargs):
"""Indicator: Squeeze Momentum (SQZ)"""
# Validate arguments
bb_length = int(bb_length) if bb_length and bb_length > 0 else 20
bb_std = float(bb_std) if bb_std and bb_std > 0 else 2.0
kc_length = int(kc_length) if kc_length and kc_length > 0 else 20
kc_scalar = float(kc_scalar) if kc_scalar and kc_scalar > 0 else 1.5
mom_length = int(mom_length) if mom_length and mom_length > 0 else 12
mom_smooth = int(mom_smooth) if mom_smooth and mom_smooth > 0 else 6
_length = max(bb_length, kc_length, mom_length, mom_smooth)
high = verify_series(high, _length)
low = verify_series(low, _length)
close = verify_series(close, _length)
offset = get_offset(offset)
if high is None or low is None or close is None: return
use_tr = kwargs.setdefault("tr", True)
asint = kwargs.pop("asint", True)
detailed = kwargs.pop("detailed", False)
lazybear = kwargs.pop("lazybear", False)
mamode = mamode if isinstance(mamode, str) else "sma"
def simplify_columns(df, n=3):
df.columns = df.columns.str.lower()
return [c.split("_")[0][n - 1:n] for c in df.columns]
# Calculate Result
bbd = bbands(close, length=bb_length, std=bb_std, mamode=mamode)
kch = kc(high, low, close, length=kc_length, scalar=kc_scalar, mamode=mamode, tr=use_tr)
# Simplify KC and BBAND column names for dynamic access
bbd.columns = simplify_columns(bbd)
kch.columns = simplify_columns(kch)
if lazybear:
highest_high = high.rolling(kc_length).max()
lowest_low = low.rolling(kc_length).min()
avg_ = 0.25 * (highest_high + lowest_low) + 0.5 * kch.b
squeeze = linreg(close - avg_, length=kc_length)
else:
momo = mom(close, length=mom_length)
if mamode.lower() == "ema":
squeeze = ema(momo, length=mom_smooth)
else: # "sma"
squeeze = sma(momo, length=mom_smooth)
# Classify Squeezes
squeeze_on = (bbd.l > kch.l) & (bbd.u < kch.u)
squeeze_off = (bbd.l < kch.l) & (bbd.u > kch.u)
no_squeeze = ~squeeze_on & ~squeeze_off
# Offset
if offset != 0:
squeeze = squeeze.shift(offset)
squeeze_on = squeeze_on.shift(offset)
squeeze_off = squeeze_off.shift(offset)
no_squeeze = no_squeeze.shift(offset)
# Handle fills
if "fillna" in kwargs:
squeeze.fillna(kwargs["fillna"], inplace=True)
squeeze_on.fillna(kwargs["fillna"], inplace=True)
squeeze_off.fillna(kwargs["fillna"], inplace=True)
no_squeeze.fillna(kwargs["fillna"], inplace=True)
if "fill_method" in kwargs:
squeeze.fillna(method=kwargs["fill_method"], inplace=True)
squeeze_on.fillna(method=kwargs["fill_method"], inplace=True)
squeeze_off.fillna(method=kwargs["fill_method"], inplace=True)
no_squeeze.fillna(method=kwargs["fill_method"], inplace=True)
# Name and Categorize it
_props = "" if use_tr else "hlr"
_props += f"_{bb_length}_{bb_std}_{kc_length}_{kc_scalar}"
_props += "_LB" if lazybear else ""
squeeze.name = f"SQZ{_props}"
data = {
squeeze.name: squeeze,
f"SQZ_ON": squeeze_on.astype(int) if asint else squeeze_on,
f"SQZ_OFF": squeeze_off.astype(int) if asint else squeeze_off,
f"SQZ_NO": no_squeeze.astype(int) if asint else no_squeeze,
}
df = DataFrame(data)
df.name = squeeze.name
df.category = squeeze.category = "momentum"
# Detailed Squeeze Series
if detailed:
pos_squeeze = squeeze[squeeze >= 0]
neg_squeeze = squeeze[squeeze < 0]
pos_inc, pos_dec = unsigned_differences(pos_squeeze, asint=True)
neg_inc, neg_dec = unsigned_differences(neg_squeeze, asint=True)
pos_inc *= squeeze
pos_dec *= squeeze
neg_dec *= squeeze
neg_inc *= squeeze
pos_inc.replace(0, npNaN, inplace=True)
pos_dec.replace(0, npNaN, inplace=True)
neg_dec.replace(0, npNaN, inplace=True)
neg_inc.replace(0, npNaN, inplace=True)
sqz_inc = squeeze * increasing(squeeze)
sqz_dec = squeeze * decreasing(squeeze)
sqz_inc.replace(0, npNaN, inplace=True)
sqz_dec.replace(0, npNaN, inplace=True)
# Handle fills
if "fillna" in kwargs:
sqz_inc.fillna(kwargs["fillna"], inplace=True)
sqz_dec.fillna(kwargs["fillna"], inplace=True)
pos_inc.fillna(kwargs["fillna"], inplace=True)
pos_dec.fillna(kwargs["fillna"], inplace=True)
neg_dec.fillna(kwargs["fillna"], inplace=True)
neg_inc.fillna(kwargs["fillna"], inplace=True)
if "fill_method" in kwargs:
sqz_inc.fillna(method=kwargs["fill_method"], inplace=True)
sqz_dec.fillna(method=kwargs["fill_method"], inplace=True)
pos_inc.fillna(method=kwargs["fill_method"], inplace=True)
pos_dec.fillna(method=kwargs["fill_method"], inplace=True)
neg_dec.fillna(method=kwargs["fill_method"], inplace=True)
neg_inc.fillna(method=kwargs["fill_method"], inplace=True)
df[f"SQZ_INC"] = sqz_inc
df[f"SQZ_DEC"] = sqz_dec
df[f"SQZ_PINC"] = pos_inc
df[f"SQZ_PDEC"] = pos_dec
df[f"SQZ_NDEC"] = neg_dec
df[f"SQZ_NINC"] = neg_inc
return df
squeeze.__doc__ = \
"""Squeeze (SQZ)
The default is based on John Carter's "TTM Squeeze" indicator, as discussed
in his book "Mastering the Trade" (chapter 11). The Squeeze indicator attempts
to capture the relationship between two studies: Bollinger Bands® and Keltner's
Channels. When the volatility increases, so does the distance between the bands,
conversely, when the volatility declines, the distance also decreases. It finds
sections of the Bollinger Bands® study which fall inside the Keltner's Channels.
Sources:
https://tradestation.tradingappstore.com/products/TTMSqueeze
https://www.tradingview.com/scripts/lazybear/
https://tlc.thinkorswim.com/center/reference/Tech-Indicators/studies-library/T-U/TTM-Squeeze
Calculation:
Default Inputs:
bb_length=20, bb_std=2, kc_length=20, kc_scalar=1.5, mom_length=12,
mom_smooth=12, tr=True, lazybear=False,
BB = Bollinger Bands
KC = Keltner Channels
MOM = Momentum
SMA = Simple Moving Average
EMA = Exponential Moving Average
TR = True Range
RANGE = TR(high, low, close) if using_tr else high - low
BB_LOW, BB_MID, BB_HIGH = BB(close, bb_length, std=bb_std)
KC_LOW, KC_MID, KC_HIGH = KC(high, low, close, kc_length, kc_scalar, TR)
if lazybear:
HH = high.rolling(kc_length).max()
LL = low.rolling(kc_length).min()
AVG = 0.25 * (HH + LL) + 0.5 * KC_MID
SQZ = linreg(close - AVG, kc_length)
else:
MOMO = MOM(close, mom_length)
if mamode == "ema":
SQZ = EMA(MOMO, mom_smooth)
else:
SQZ = EMA(momo, mom_smooth)
SQZ_ON = (BB_LOW > KC_LOW) and (BB_HIGH < KC_HIGH)
SQZ_OFF = (BB_LOW < KC_LOW) and (BB_HIGH > KC_HIGH)
NO_SQZ = !SQZ_ON and !SQZ_OFF
Args:
high (pd.Series): Series of 'high's
low (pd.Series): Series of 'low's
close (pd.Series): Series of 'close's
bb_length (int): Bollinger Bands period. Default: 20
bb_std (float): Bollinger Bands Std. Dev. Default: 2
kc_length (int): Keltner Channel period. Default: 20
kc_scalar (float): Keltner Channel scalar. Default: 1.5
mom_length (int): Momentum Period. Default: 12
mom_smooth (int): Smoothing Period of Momentum. Default: 6
mamode (str): Only "ema" or "sma". Default: "sma"
offset (int): How many periods to offset the result. Default: 0
Kwargs:
tr (value, optional): Use True Range for Keltner Channels. Default: True
asint (value, optional): Use integers instead of bool. Default: True
mamode (value, optional): Which MA to use. Default: "sma"
lazybear (value, optional): Use LazyBear's TradingView implementation.
Default: False
detailed (value, optional): Return additional variations of SQZ for
visualization. Default: False
fillna (value, optional): pd.DataFrame.fillna(value)
fill_method (value, optional): Type of fill method
Returns:
pd.DataFrame: SQZ, SQZ_ON, SQZ_OFF, NO_SQZ columns by default. More
detailed columns if 'detailed' kwarg is True.
"""