Pair Trading - Exploring The Low Risk Statistical Arbitrage Trading Concepts

[VJAY]

Not sure, I am at office now will check in the evening, but this pair is output in my system. Please check if these 2 pairs have any split/bonus in the last 200 days, in which case pValue would be impacted. You can quickly check for splits by plotting the pairs as follows:
returns[['IL&FSTRANS','ITDCEM']].plot()

 

[ncube]

View attachment 29108

Can you also plot df[['IL&FSTRANS','ITDCEM']].plot()

 

[VJAY]

 

[ncube]

yes the image is for same

no you have plotted returns[], change it to df[]

 

[VJAY]

no you have plotted returns[], change it to df[]

yes edited post

 

[ncube]

View attachment 29125

Ok..Looks to be fine, not sure why the pvalue is different, one of the issues of using machine learning, few things become blackbox. Anyway let me check this in the evening.

 

[VJAY]

Ok..Looks to be fine, not sure why the pvalue is different, one of the issues of using machine learning, few things become blackbox. Anyway let me check this in the evening.

if today traded then buy triggered in ITDCEM...short not triggered

 

[ncube]

if today traded then buy triggered in ITDCEM...short not triggered

Ok anyway we can track this as the pair setup is good.

 

[ncube]

Dear ncube,p value of this pair is not more ?

View attachment 29103

Seems pvalue is correct in my script, its about 0.0015, check how many days data you passing in S1,S2, I have restricted it to 200 days data.

The zscore today is 2.614, so we need to take second increment of trades tomorrow as follows:

 

[VJAY]

Seems pvalue is correct in my script, its about 0.0015, check how many days data you passing in S1,S2, I have restricted it to 200 days data.
View attachment 29135
The zscore today is 2.614, so we need to take second increment of trades tomorrow as follows:
View attachment 29136

I had not changed anything in your code which shared with us ....