Computing daily value weighted average portfolio returns

[ties jongen]

I am trying to compute value weighted average daily portfolio returns for a portolio, but am not sure how to correctly code this

my data looks the following:

date	company	companyreturn	marketcap2016	marketreturn	industry	eventdummy	Market weighted return (not yet created)
1	x	1	300	3	d	0	UU
2	x	2	300	2	d	1	VV
3	x	1	300	3	d	0	WW
1	y	0	200	3	e	0	UU
2	y	3	200	2	e	1	VV
3	y	1	200	3	e	0	WW
1	z	-1	150	3	f	0	UU
2	z	-3	150	2	f	1	VV
3	z	5	150	3	f	0	WW

How can I create the last column "Market weighted return (not yet created)"

Thereafter, I would like to reduce the table size to the following

date	marketreturn	eventdummy	Market weighted return (not yet created)
1	3	0	UU
2	2	1	VV
3	3	0	WW

This i expect to do the following:

duplicates drop date marketreturn eventdummy marketweightedreturn
drop company companyreturn marketcap2016 industry.


I am just a bit lost on how to first create the market weighted return (that consists of the daily value(market cap) weighted average return for the portfolio that consists of all the companies in the sample


Could you help me with this?

[Joro Kolev]

I am not sure how to read the data that you are showing -- what are those returns that are 1,3,5, -3, etc...

But you can calculate weighted averages with my function -egen, wmean()-

From within Stata type

 ssc install _gwmean

and then to calculate the weighted average of return for each date:

egen meanret = wmean(ret), by(date) weights(market_cap)

[ties jongen]

The returns are just an example; in my real data they are shown as decimals eg; 0.0324645 being around 3.2%


I will look at it now, thank you for the fast response, again! :D

[ties jongen]

It worked, thank you so much for the help!