Exponentially Weighted Moving Average Using Pandas

Question

I need to confirm a few things related to the exponential weighted moving average function for pandas.

If I have a df dataset for which I need to find a 12-day exponential moving average, will there be a correct method below.

exp_12=df.ewm(span=20,min_period=12,adjust=False).mean()

Given that the data set contains 20 readings, the range (Total number of values) should be 20.

Since I need to find a moving average of 12 days, therefore, min_period = 12. I interpret the interval as the total number of values ​​in the data set or the total covered time.

Can anyone confirm the correctness of my interpretation? I can not understand the meaning of the adjustment.

I have attached a link to the pandas.df.ewm documentation below.

http://pandas.pydata.org/pandas-docs/stable/generated/pandas.DataFrame.ewm.html

Answer 1

A quote from Pandas docs : Span corresponds to what is commonly called the "EW N-day moving average."

In your case, set span = 12. You do not need to indicate that you have 20 data points, the panda will take care of this. min_period may not be required here.

Answer 2

I'm trying to play a script, doing in my case

msft_ewma['EWMA26'] = msft_ewma.ewm(span=26,min_periods=26,adjust=False).mean()

and brings me back

ValueError: Wrong number of items passed 2, placement implies 1

What for?