How do i use the & to add two booleans?-CodePudding

can anyone tell me why im getting:

TypeError: unsupported operand type(s) for &: 'float' and 'bool'

on this:

  quotesstock['crossup'] = (quotesstock['volatility']*5.5 > quotesstock['vix'] & quotesstock['volatility'].shift(-1, axis = 0) <= quotesstock['vix'].shift(-1, axis = 0))

the individual parts do give a lists of dates(index) and booleans but i cant seem to implement an and.

quotesstock['volatility']*5.5 > quotesstock['vix']
Out[158]: 
Date
2019-12-31    False
2020-01-02    False
2020-01-03    False
2020-01-06    False
2020-01-07    False
 
2022-04-19     True
2022-04-20     True
2022-04-21     True
2022-04-22    False
2022-04-25    False
Length: 584, dtype: bool

quotesstock['volatility'].shift(-1, axis = 0) <= quotesstock['vix'].shift(-1, axis = 0)
Out[159]: 
Date
2019-12-31    False
2020-01-02    False
2020-01-03    False
2020-01-06    False
2020-01-07    False
 
2022-04-19     True
2022-04-20     True
2022-04-21     True
2022-04-22     True
2022-04-25    False
Length: 584, dtype: bool

but

quotesstock['crossup'] = (quotesstock['volatility']*5.5 > quotesstock['vix'] & quotesstock['volatility'].shift(-1, axis = 0) <= quotesstock['vix'].shift(-1, axis = 0))
Traceback (most recent call last):

  File "D:\MachineLearning\Anaconda\lib\site-packages\pandas\core\ops\array_ops.py", line 302, in na_logical_op
    result = op(x, y)

TypeError: unsupported operand type(s) for &: 'float' and 'bool'

CodePudding user response：

This is due to Operator precedence, & is more binding (more sticky) than <, <=, >, >=, !=, == so for example

1.5 < 2.5 & True

is understood as

1.5 < (2.5 & True)

which cause

TypeError: unsupported operand type(s) for &: 'float' and 'bool'

you need to use brackets to get < comparison evaluted first, in above example that mean you need to do

(1.5 < 2.5) & True

in order to get True