Create a Var by selecting from a List

November 11, 2023

i want to create a Var that is conditional on an observation being in a list and some other criteria, so i tried this

emp_list1 = ["Gov.Sector","Semi Gov.Sector"]
PF_deftag['emp_grp1'] = np.where(
    (PF_deftag['Employement Sector'] in ['emp_list1'] )
    & (PF_deftag['Monthly salary in hand'].astype(float) < 3000),
    1,
    0
)

but get an error

ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().

In this case, no worries as this works

PF_deftag['emp_grp1'] = np.where(
    ((PF_deftag['Employement Sector'] == "Gov.Sector") | 
        (PF_deftag['Employement Sector'] == "Semi Gov.Sector"))
    & (PF_deftag['Monthly salary in hand'].astype(float) < 3000),
    1,
    0
)

but the next condition may have 6 or 7 components to the list?

thanks

>Solution :

import numpy as np

# List of employment sectors you're interested in
emp_list1 = ["Gov.Sector", "Semi Gov.Sector"]

# Check if each row's 'Employement Sector' is in the list and monthly 
salary is less than 3000
PF_deftag['emp_grp1'] = np.where(
    (PF_deftag['Employement Sector'].isin(emp_list1)) & 
(PF_deftag['Monthly salary in hand'].astype(float) < 3000),
    1,
    0
)

Explanation:

Import NumPy: import numpy as np – Import the NumPy library for numerical operations.

Define Employment Sectors: emp_list1 = ["Gov.Sector", "Semi Gov.Sector"] – Create a list of employment sectors you want to consider.

Apply Conditions: PF_deftag['emp_grp1'] = np.where(...) – Check if each row’s ‘Employement Sector’ is in the specified list and if the monthly salary is less than 3000. Assign 1 if true, 0 if false, to a new column ’emp_grp1′.

And for additional lists, you can expand it like this:

emp_list1 = ["Gov.Sector", "Semi Gov.Sector"]
emp_list2 = ["SomeSector1", "SomeSector2"]

PF_deftag['emp_grp1'] = np.where(
    (
        (PF_deftag['Employement Sector'].isin(emp_list1) | 
PF_deftag['Employement Sector'].isin(emp_list2))
    ) & (PF_deftag['Monthly salary in hand'].astype(float) < 3000),
    1,
    0
)