Search results for query: *

F.Y.I. DBDesigner4 is going to be superseded by MySQL Wokbench (http://www.mysql.com/products/tools/workbench/). It is still in alpha and buggy.

you could use sql, e.g. something like: proc sql; create table blar as select t1.* ,case when (select count(*) from table t2 where t2.syndrome = t1.syndrome and t2.patientclass = t1.patientclass...

you could also try set difference... proc sql; create table blar as select * from table_1 except select * from table_2; quit; and vice versa cheers, dan.

Thanks for the info Chris - i'll check it out. Cheers, Dan.

ok, so the problem is memory not IO. i had a look at proc corr using "options fullstimer" & saw that its memory requirement seems quadratic to the number of varibles. in order to select a unique list of companyies: proc sql; create table company as select distinct permno from mydataset...

1)how many records? how many different companies? 2)can you show us the code that you are using and tell us which bit of the code is problematic ie "too slow"? 3) are you getting any errors, if so can you show us the log. there are some things you can do to improve performance. e.g. using a...

I'm not sure if there is a direct method, but SAS help has examples of exporting to a specific sheet and exporting based on a criteria. So, you could adapt the examples like i've done below, ie get a unique list of values that use are segmenting by (in this example its 'M' & 'F' (sex)) and store...

i rearly use 'PROC TRANSPOSE' but did your first proc transpose work? i think you need to change it to: proc transpose data=all out=tran_all prefix=ret; by date; id permno; /* i don't think the next line is neccessary*/ /*var ret;*/ run; then change your next proc transpose to (suppress...

yes, it should work on your data - you just need to change the variable names in the code i gave you. in my code 'stock' is the variable that uniquely identifies a company - so that is 'COMNAM' for you. I've also assumed that one record uniquely identifies one stock quote for a company on one...

we need to know how your current data is structured. another way (as opposed to transposing) is to create a cross product of stock quotes by day then use 'PROC CORR' with a 'BY' statement to produce the correlations e.g. i assumed your data looks something like: stock date quote AAA...

well, i got a little curious about performance so i implemented method 2) above, ie using a hash table to implement the ADT. The hash table actually stores a set of trees where each tree represents a set of equivalent customers. The performance is very much dependant on the number of customers...

how many customers? 1) you could load your customer reference table into a hash table such that you can use (near) direct addressing instead of 'searching'. 2) implement equivalant relations (disjoint sets ADT) using hash tables. i can explain more if these methods are appropriate for you...

Hi, Im Dan from New Zealand. I've been using SAS for about 4 years - am SAS Base Certified. I am a data analysis within the injury prevention field, but i'm more of a database person. I have agricultural science, epidemiological (statistical) and computer science tertiary education. I programme...

F.Y.I. i recently looked into exporting from SAS to STATA & SPSS and found that, on top of what you have already discovered with SPSS v14, SAS v9.1.3 has also added some data sources to its PROC EXPORT procedure. Have a look at the 9.1.3 SAS On-line Doc e.g. data test; a=10; b=20; output...

Given your suituation requires consideration to the finer aspects of securing an MS database/app, then i recommend you purchasing: "Real World Microsoft Access Database Protection and Security" It's very thorough, and even discusses things like setting up folder permissions such that users...

you could put a password on the backup db? e.g. a 20 character password that only you know - your users won't know it. The client app will have the password in the code, but if you put the password in a module and turn the app into a mde it should be fine for in-house "security". would that be...

Hi John, Glade to hear the software was useful. what would also be really cool is, if the software added some functionality (e.g. as a plug-in) for infering functional dependencies in the presence of messy data. I did a google search and found this application/code. I haven't tried it though...

some remarks about adp project support for access 2007 is mentioned in this blog. It seems that MS recommend linked tables instead(??) In terms of a developers edition, then i think you need Visual Studio Tools for Office (VSTO) for the license to ship a run-time license. This seems to be true...

just to add a couple of open source tools to the list: - DBDesigner4 - DBDesigner Fork cheers, dan. PS Good luck!

(an aside) question: How come you need to decrypt? Wouldn't you be better of using a one-way encryption algorithm - the password is validated by an exact string match on the encrypted strings ie Does the encryped user-supplied password match the stored value. cheers, dan.