That's ok art297, many thanks for your help! Maybe SAS calculates the avarege of the whole time series and substracts it from the observations instead of calculating each month's seperetaly. I will try to find out why I get different results. Many thanks again ... View more

I tried both ways but I still get diferrent reasults. Maybe it is the methodology of the paper that I am using different. What I need to do is: let Pt the price of stock today. 1. convert prices: [log(Pt)-log(Pt-1)]*100 (let's call it X) 2.get the average of each month (lets call it x) 3. substract each X by the average. Xt-x (let's call it y) 4.miltiply y*y. 5.Finally get the average of the above for each month to get the monthly volatility. ... View more

Something else, I compared the calculations from SAS with those that I did manually for some stocks and they have totally different results. Maybe the model I use measure s differently the volatility. So, if I want to to do what I said above, I mean from the daily prices of each month substract the average of the specific month how I would do it? ... View more

It works!!! The only problem now is than in my output table  (the one I want to export) there is no USS, only min, max, mean, std and N. How can i fix this? Thanks ... View more

The date has the form of   21JUN1996 (format: DATE9.) and the entries have the form 1.0234345595 (fomat: BEST12.) ... View more

Sorry for bothering you again but I am really newbie into SAS and I don't understand how some things work. I imported my data from an excel file which has 100 stocks and some thousands of observations. Normally I should skip the cards part that u have written. I am a bit confused with the yrmon varibale since the data is imported through the manager and already exists as a table, If I write data have; input DATE A1-A100; proc means data=have mean uss; var a1--a100; run; I get the overall USS for each variable, not the monthly, I tried with yrmon=put(DATE,yymmd.); after the input statement and later in the proc means statement as u wrote but it doesn't work. How I could modified this in order to make it work? I just want to produce a file I can export with the desired results. Sorry If I make any silly questions but as I said I am really new into SAS. thanks for your understanding ... View more

Hi art297, thanks for your reply. It should be easy, though I am new to SAS and I do not have sufficient knowledge on it. So far I have a table with the stocks' log diferrences (100 stocks) and another table with the monthly averages of them. What I want to do is to create a new table where the variables will be the log price minus the monthly average. Keep in mind tha the log prices are daily and the average is for each month, so all prices of january 2000 will be minus the average of this month, the prices of february 2000 with the average of this month and so on. Below is how the data looks like: My log prices table: DATE A1 A2 A100 31/1/2000 0.256 0.5424 0.564 1/2/2000 0.321 0.1223 0.446 2/2/2000 0.453 0.41554 0.265 3/2/2000 0.1524 0.151 0.456 4/2/2000 0.2045 0.1656 0.355 My average table has prices of the form: DATE A1 A2 A100 JAN2000 0.152 0.156 0.256 FEB2000 0.235 0.245 0.3565 MAR2000 0.456 0.654 0.265 What I want to get is : DATE A1 A2 A100 31/1/2000 0.256-0.152 0.5424-0.156 0.564-0.256 1/2/2000 0.321-0.235 0.151-0.245 0.456-0.3565 2/2/2000 0.453-0.235 0.1656-0.245 0.355-0.3565 (numbers only for reference) So each price will be substracted by the given average of the specific month. Any suggestions? ... View more

Yeap, got the mistake Thank you all ... View more

Guys I tried all the proposed sollutions but I am facing problems, In the first case with sql I get the following prob: 52 proc sql; 53 54 create table stock3 as 55 56 select 57 58 intnx('month',date,0,'end') as date format ddmmyys8., 59 60 sum(a1) as a1, 61 62 sum(a2) as a2, 63 64 sum(a3) as a3 65 66 from 67 68 vol 69 70 group by 71 72 calculated date; ERROR: The SUM summary function requires a numeric argument. 73 74 quit; NOTE: The SAS System stopped processing this step because of errors. NOTE: PROCEDURE SQL used (Total process time): real time 0.01 seconds cpu time 0.01 seconds Whereas using the proc means I get: proc means noprint chartype nway missing data=vol; 26 class date; 27 var a1-a98; ERROR: Variable a1 in list does not match type prescribed for this list. 28 format date monyy6.; 29 output out=vol2 (drop=_type__freq_)sum=; 30 run; NOTE: The SAS System stopped processing this step because of errors. WARNING: The data set WORK.VOL2 may be incomplete. When this step was stopped there were 0 observations and 0 variables. NOTE: PROCEDURE MEANS used (Total process time): real time 0.01 seconds cpu time 0.01 seconds Any suggestions? I am attaching a sample file ... View more

It worked! Many thanks tom ... View more

Many thanks guys! NickR it works great! You saved me lots of time. ... View more