Pandas: AttributeError: ‘float’ object has no attribute ‘MACD’

I would like to compare 2 rows in a pandas dataframe but I always get an Error saying: AttributeError: ‘float’ object has no attribute ‘MACD’.

This is the df:

time    open    high    low close   tick_volume spread  real_volume EMA_LONG    EMA_SHORT   MACD    SIGNAL  HIST    200EMA
0   2018-01-05 03:00:00 1.20775 1.20794 1.20700 1.20724 2887    1   0   1.206134    1.206803    0.000669    0.000669    0.000000    1.207240
1   2018-01-05 04:00:00 1.20723 1.20743 1.20680 1.20710 2349    1   0   1.206216    1.206849    0.000633    0.000649    -0.000016   1.207170
2   2018-01-05 05:00:00 1.20709 1.20755 1.20709 1.20744 1869    1   0   1.206318    1.206941    0.000622    0.000638    -0.000016   1.207261

Now I want to count on how many times it would buy and sell based on some information in the rows so I’m trying to iterate through it like this:

buy = 0
sell = 0

for i, row in df.iterrows():
    if i == 0:
        continue
    if row.MACD > row.SIGNAL and row[i - 1].MACD < row[i - 1].SIGNAL:
        if row.HIST < 0 and row.MACD > row['200EMA'] and row.SIGNAL > row['200EMA']:
            buy += 1
    elif row.MACD < row.SIGNAL and row[i - 1].MACD > row[i - 1].SIGNAL:
        if row.HIST > 0 and row.MACD < row['200EMA'] and row.SIGNAL < row['200EMA']:
            sell += 1

print("BUY: " + buy + "SELL: " + sell)

I am getting the following Error:

AttributeError: 'float' object has no attribute 'MACD'
---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-75-ff4a2b3629bc> in <module>
      8         if row.HIST < 0 and row.MACD > row['200EMA'] and row.SIGNAL > row['200EMA']:
      9             buy += 1

---> 10     elif row.MACD < row.SIGNAL and row[i - 1].MACD > row[i - 1].SIGNAL:
     11         if row.HIST > 0 and row.MACD < row['200EMA'] and row.SIGNAL < row['200EMA']:
     12             sell += 1

AttributeError: 'float' object has no attribute 'MACD'

I know this Error has already been here but I the solutions there didn’t help me.

Thank you already!

Answer

your problem his here row[i – 1].MACD

when you are accesessing the row[i-1] place you get the value of the location in the service if i = 1 then you will get the row[0] for the row and not the preivice row in the dataframe you should probably switch it by df.iloc[i-1].MACD