Autocorrelation: what exactly 1 month lag mean? 30 days lag? same day of month?

Installations & Python Packages
1

Course Name: Financial Time Series Analysis for Trading, Section No: 11, Unit No: 11, Unit type: Notebook

# Plot ACF of wheat ETF for past 40 lags
plot_acf(wheat_price.Close, lags=40)

Can anyone tell me what exactly is a month for this autocorrelation plot?
How it is handles when is was a not trading day 1 month ago?
Let's say I want to apply this for trading signal, it is important to be precise with the method of taking a lag
 

2

cancel that, question is no longer irrelevant.



autocorrelation in financial time series is pretty useless

check this video if you want to learn more why 

3

Hi Dmitry,



So basically, there can be some instances when the returns of a financial time series can be auto-correlated. Though this might not occur frequently, if such an auto-correlation exists, you can take advantage of this alpha to trade the asset. For ex. - if the autocorrelation of returns with a lag of 1 month is significantly positive and returns are positive for this month, then there is a good probability that the returns might be positive for next month.



Hope this helps!



Thanks,

Akshay