cancel
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for 
Search instead for 
Did you mean: 
SAM_22
‎03-30-2019
  • 5 Posts
  • 3 Likes Given
  • 0 Solutions
  • 0 Likes Received

Re: Construct portfolio by adding stocks daily and dropping them after...

by Fluorite | Level 6 in New SAS User
‎03-30-2019 03:45 AM
‎03-30-2019 03:45 AM
Thanks for replying. I attached my data sample: I just want to average returns daily. day 1: average returns in day 1 day 2: average returns in day 1&2; day 3: average returns in day 1,2&3; but after being hold for a month, the stocks must be dropped, just like the way i explained before. It doesnt need to be exact jan 1 to feb 1 it can be interval like 20 or 30 days   ... View more

Construct portfolio by adding stocks daily and dropping them after cer...

by Fluorite | Level 6 in New SAS User
‎03-29-2019 02:28 PM
‎03-29-2019 02:28 PM
Hello, i would like to form transaction based calendar time portfolios. The portfolios are formed by mimicking daily trades (adding and dropping stocks after certain holding period, e.g. a month). I have trading data with many buy and sell trades in a day. If i understand it correctly, this is how daily portfolios are constructed:   Day 1 (Jan 1): calculate the average return of stocks bought by investors in the first day. Day 2 (Jan 2): calculate the average return of stocks bought in day 1&2. Day 3 (Jan 3): calculate the average return of stocks bought in day 1, 2 & 3. Day 4 (Jan 4): and so on. After a month (e.g. Feb 1), all the stocks from day 1 must be dropped, while adding new stocks from today's trade. On the next day, stocks from Feb 2 are dropped, while adding new ones from this day. On Feb 3, stocks from Jan 3 are dropped, while adding new ones from this day. and so on;   The results should be like this;   Date Portfolios Stocks bought portfolio return Jan-01 Portfolio day 1 stock 1, stock 2, stock 3 average return (stock 1,2&3) Jan-02 Portfolio day 2 stock 4, stock 5, stock 6 average return (stock 1,2,3,4,5,&6) Jan-03 Portfolio day 3 stock 7, stock 8, stock 9 average return (stock 1,2,3,4,5,6,7,8&9)   .   continue wih the same pattern   .   .   .   . After one month (assuming 20 trading days in a month)         Feb-01 Portfolio day 21 (assuming  20 trading days in a month)  stock 10, stock 11, stock 12 average return (stock 4,5,6,7,8,9,10,11&12) stocks 1,2,&3 are dropped Feb-02 Portfolio day 22 stock 13, stock 14, stock 15 average return (stock 7,8,9,10,11,12,13,14&15) stocks 4,5,&6 are dropped Feb-03 Portfolio day 23 stock 16, stock 17, stock 18 average return (stock 10,11,12,13,14,15,16,17,&18) stocks 7,8,&9 are dropped       and so on   So far i only know the codes for regular moving average, and do not know how to do it like this. Any help would be greatly appreciated. Thank you in Advance!   ... View more

Re: Count non-missing value (trading days) every month and remove thos...

by Fluorite | Level 6 in New SAS User
‎12-17-2018 03:05 PM
‎12-17-2018 03:05 PM
It's actually hundreds of stocks, i just posted the sample as an illustration. I've followed your code and it does exactly what i need. Thank you so much! ... View more

Re: Count non-missing value (trading days) every month and remove thos...

by Fluorite | Level 6 in New SAS User
‎12-17-2018 01:26 PM
‎12-17-2018 01:26 PM
Hi Reeza, I'll look into it, thank you! ... View more

Count non-missing value (trading days) every month and remove those wi...

by Fluorite | Level 6 in New SAS User
‎12-17-2018 01:13 PM
‎12-17-2018 01:13 PM
Hello I'm new to SAS and currently working on stock price data. Here is the example of my dataset:   Date Stock1 Stock2 Stock3 Stock4 Stock5 Stock6 Stock7 Stock8 01/06/2000 314,11 212,49 165 450 98,29   125,51 185 02/06/2000 314,11 212,49 143 425 95,83   115,85 185 05/06/2000 338,27 212,49 136 450 98,29   123,1 185 06/06/2000 338,27 212,49 131 450 100,75   125,51 200 07/06/2000 314,11 212,49 133 475 100,75   135,16 200 08/06/2000 362,43 212,49   525 108,12   135,16 172,5 09/06/2000 338,27 212,49   475 103,2   139,99 200 12/06/2000 338,27 212,49   500 100,75   132,75 200 13/06/2000 314,11 212,49   500 100,75   130,34 200 14/06/2000 314,11 212,49   500 103,2   135,16 200 15/06/2000 314,11 212,49   500 103,2   135,16 200 16/06/2000 338,27 212,49   500 108,12   142,4 200 19/06/2000 362,43 212,49   550 108,12   147,23 217,5 20/06/2000 338,27 212,49   525 120,4   144,82 217,5 21/06/2000 362,43 212,49   550 125,32   149,65 217,5 22/06/2000 362,43 212,49   600 125,32   164,13 200 23/06/2000 338,27 212,49   600 122,86   161,71 200 26/06/2000 386,6 212,49   575 120,4   161,71 210 27/06/2000 338,27 212,49   575 117,95   161,71 190 28/06/2000 338,27 212,49   575 113,03   161,71 190 29/06/2000 338,27 212,49   550 113,03   164,13 190 30/06/2000 338,27 212,49   575 120,4   168,95 190 03/07/2000 338,27 212,49   550 112,05   160,75 190 04/07/2000 338,27 212,49   565 117,95   159,3 190 05/07/2000 347,94 212,49   540 123,85   159,3 190 06/07/2000 347,94 212,49   555 126,3   163,64 180,5 What i want to do is to count non-missing values (trading days) every month for every variable (stock 1 to stock 😎 and create new dataset that keep variables with minimum 15 trading days every month, and remove those with less than 15 trading days every month. Based on the dataset above, i will keep variable 1,2,4,5,7, & 8, and drop variable 3 and 6.   Or maybe just tell me how to count the non-missing value every month and i will remove the variable manually. I have read some manuals but still couldn't find the solution. Any help/suggestion will be greatly appreciated. Thank you so much ... View more
Contact Me
Online Status
Offline
Date Last Visited
‎03-30-2019 12:04 PM
Likes given to
View All
Quick links