How to add new columns to pandas data frame using .loc - pandas

Question

I am calculating very simple daily stock calculations in data frame ( for e.g. SMA, VWAP, RSI etc). After I upgraded to anaconda 3.0, my code stopped working and gives followed error. I don't have much experience in coding and need some help.
KeyError: "Passing list-likes to .loc or [] with any missing labels is no longer supported. The following labels were missing: Index(['RSI', 'ZONE'], dtype='object'). See https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#deprecate-loc-reindex-listlike"
Followed is the code.
import yfinance as yf
import pandas as pd
def convert_to_dataframe_daily(data):
window = 10
window20 = 20
window50 = 50
window100 = 100
window200 = 200
ema_time = 8
#data = yf.download("googl", period="30d", interval="5m")
#data = yf.download('TSLA', period='30d', interval='5m')
pd.set_option('display.max_columns', None)
#calculation for VWAP
volumeC = data['Volume']
priceC = data['Close']
df = data.assign(VWAP=((volumeC * priceC).cumsum() / volumeC.cumsum()).ffill())
#Convert the timezone to Chicago central
#df.index = pd.DatetimeIndex(df.index.tz_convert('US/Central')) # aware--> aware
#reset the dataframe index and separate time
df.reset_index(inplace=True)
#df.index.intersection
#df2 = df[df.index.isin(dts)]
#df['Date'] = pd.to_datetime(df['Datetime']).dt.date
#df['Time'] = pd.to_datetime(df['Datetime']).dt.time
# calculate stochastic
df['low5']= df['Low'].rolling(5).min()
df['high5']= df['High'].rolling(5).max()
#k = 100 * (c - l) / (h - l)
df['K'] = (df['Close']-df['low5'])/(df['high5']-df['low5'])
#s.reindex([1, 2, 3])
columns = df.columns.values.tolist()
#df[columns[index]]
#df = pd.DataFrame(np.random.randn(8, 4),index=dates, columns=['A', 'B', 'C', 'D'])
df = df.loc[:, ('Date','Open','High','Low', 'Close','Volume','VWAP','K','RSI', 'ZONE')]
#df = df.reindex(['Date','Open','High','Low', 'Close','Volume','VWAP','K','RSI', 'ZONE'])
df['RSI'] = calculate_rsi(df)
filter_Z1 = df['K'] <=0.1
filter_Z2 = (df['K'] > 0.1) & (df['K'] <= 0.2)
filter_Z3 = (df['K'] > 0.2) & (df['K'] <= 0.3)
filter_Z4 = (df['K'] > 0.3) & (df['K'] <= 0.4)
filter_Z5 = (df['K'] > 0.4) & (df['K'] <= 0.5)
filter_Z6 = (df['K'] > 0.5) & (df['K'] <= 0.6)
filter_Z7 = (df['K'] > 0.6) & (df['K'] <= 0.7)
filter_Z8 = (df['K'] > 0.7) & (df['K'] <= 0.8)
filter_Z9 = (df['K'] > 0.8) & (df['K'] <= 0.9)
filter_Z10 = (df['K'] > 0.9) & (df['K'] <= 1)
#plug in stochastic zones
df['ZONE'].where(-filter_Z1, 'Z1', inplace=True)
df['ZONE'].where(-filter_Z2, 'Z2', inplace=True)
df['ZONE'].where(-filter_Z3, 'Z3', inplace=True)
df['ZONE'].where(-filter_Z4, 'Z4', inplace=True)
df['ZONE'].where(-filter_Z5, 'Z5', inplace=True)
df['ZONE'].where(-filter_Z6, 'Z6', inplace=True)
df['ZONE'].where(-filter_Z7, 'Z7', inplace=True)
df['ZONE'].where(-filter_Z8, 'Z9', inplace=True)
df['ZONE'].where(-filter_Z9, 'Z9', inplace=True)
df['ZONE'].where(-filter_Z10, 'Z10', inplace=True)
df = df['Date','Open','High','Low', 'Close','Volume','VWAP','K','RSI', 'ZONE']
return df
data = yf.download('ba', period='500d', interval='1d')
df = convert_to_dataframe_daily(data)
print(df)

Answer 1

A few lines need to be tweaked
Instead of
df = df.loc[:, ('Date','Open','High','Low', 'Close','Volume','VWAP','K','RSI', 'ZONE')]
use
df = df[['Date','Open','High','Low', 'Close','Volume','VWAP','K']]
before
df['ZONE'].where(-filter_Z1, 'Z1', inplace=True)
...
put a line
df['ZONE'] = 0
The line before return df should be changed to
df = df[['Date','Open','High','Low', 'Close','Volume','VWAP','K','RSI', 'ZONE']]