# Keep the first column
turnover_df = pd.DataFrame(data=turnover.iloc[:, 0].values, columns=['Turnover_Date'])

QLD_cols = [col for col in turnover.columns.values if 'Queensland' in col and 'Clothing retailing' in col]
turnover_df[turnover[QLD_cols].columns] = turnover[QLD_cols]
print(turnover_df.shape)
turnover_df.head()

# Rename columns
t = turnover_df.rename(columns={turnover_df.columns[1]: 'NSW',
                               turnover_df.columns[2]: 'QLD',
                               turnover_df.columns[3]: 'VIC',
                               turnover_df.columns[4]: 'WA'})

t.head()

turnover_df.to_csv(r'./turnover.csv', index = False, header = True)