About peter_sjogarde

peter_sjogarde · ‎12-30-2016

Thank you very much for the help on this issue!

peter_sjogarde · ‎12-30-2016

Thank you Rick! Now I understand the problem and can proceed with my work. One last question; Do you know how the procedure deals with the problem? Since I get a result it has delt with the problem somehome. In the data Y={0,1,2,2,3,4,5,6,7,8}. The first decile is 0 and the second decile 1 but what about decile 3 and 4, are the values arbitrarily assigned?

peter_sjogarde · ‎12-30-2016

It is likely the case that proc quantreg cannot discriminate between quantiles. My data is skewed, have a lot of duplicates since it is derived from integers. The distribution looks like this. And zooming in to the values between 1 and 5: I have about 9.5 million observations.

peter_sjogarde · ‎12-28-2016

The smooth algorithm is not efficient enough. I stopped this procedure after about 14 hours. The ipm algorithm is not accepted. An error message says that the procedure expects interior, simplex or smooth.

peter_sjogarde · ‎12-26-2016

Thank you for the explanation. Unfortunately it did not work. I tried setting the Kappa option as low as 0.1. I guess it has to do with the skewed distribution of my data. I only get a warning for quantiles 0.1-0.5. What does the value actually say? How important is the accuracy?

peter_sjogarde · ‎12-22-2016

Thank you for your answer. I am afraid that I do not fully understand the description of the Interior point algorithm. How would reducing the Kappa option effect the results? Here is the code: proc quantreg data=t_quantreg; model cP = var1 var2 var3 var4 / quantile= 0.10 to 0.90 by 0.10 PLOT=QUANTPLOT; run;

peter_sjogarde · ‎12-20-2016

I am getting a warning message when running Proc quantreg: "WARNING: Convergence was not attained. Duality gap 4.3888576E-8 could not be reduced further. Reduce the required accuracy or try a smaller step length with the KAPPA= option." I do not understand this message. I guess that reducing the accuracy only takes the message away but does not change anything. I do not understand what reducing the step length with the KAPPA option does. Does it change the outcome or just the efficiency? Is there an easy sollution to this problem? I would be greatful if someone could explain this warning message. Is it a severe warning?

peter_sjogarde · ‎12-01-2015

Thank you! I suppressed the titles and used the ods text feature instead and that worked fine.

peter_sjogarde · ‎11-30-2015

Thank you for your reply. I posted a simplified code since the real code is quite long and complicated. The reason I want to do it this way is that my code includes a transpose part which is done separetly for each variable. If done on all variables at ones, transpose will convert number formats. orginial table looks something like this country year var1 var2 1 2001 1 0.1 1 2002 2 0.2 1 2003 1 0.5 2 2001 1 0.1 2 2002 2 0.2 2 2003 1 0.5 The program makes one table per variable and presents it like this. The macro also makes it possible for me to exclude variables if I don't want them in a particular analysis. var1 country 2001 2002 2003 1 1 2 1 2 1 2 1 var2 country 2001 2002 2003 1 0.1 0.2 0.5 2 0.1 0.2 0.5

peter_sjogarde · ‎11-30-2015

Hi, I am trying to create titles in an ODS report with multiple tables. I would like one title above all tables and then one title for each table. The table are created by iteration. The title for each table is put in a variable. Output that i want: Title 1 Title 2 [table 1] Title 3 [table 2] But there seems to be something that I do not understand about how the titles are generated since I get very strange results. Some test code: * Open ods; ods listing close; ods graphics on; ods html file="&output_folder.\&cluster_label.\&cluster_label..html" headtext="<style>p,hr {display:none}</style>" style=Journal gpath="&output_folder.\&cluster_label.\graph" image_dpi=300; ods text='<div><h1 align="center">Cars</h1></div>'; %macro car_var; data &var_name.; set sashelp.cars; if type = "&var_name."; run; Title2 "<H2>&var_title.</H2>"; proc print label data=&var_name.; run; %mend car_var; %Let var_name = Sports ; %Let var_title = Sports; %car_var %Let var_name = Sedan ; %Let var_title = Sedan; %car_var ods html close; Results: <H2>Sedan</H2> <div><h1 align="center">Cars</h1></div> Obs Make Origin 1 Audi Europe 2 Audi Europe 3 Audi Europe 4 Audi Europe 5 BMW Europe 6 BMW Europe 7 BMW Europe 8 BMW Europe 9 Jaguar Europe 10 Jaguar Europe 11 Jaguar Europe 12 Jaguar Europe 13 Mercedes-Benz Europe 14 Mercedes-Benz Europe 15 Mercedes-Benz Europe 16 Mercedes-Benz Europe 17 Mercedes-Benz Europe 18 Porsche Europe 19 Porsche Europe 20 Porsche Europe 21 Porsche Europe 22 Porsche Europe 23 Porsche Europe <H2>Sedan</H2> Obs Make Origin 1 Audi Europe 2 Audi Europe 3 Audi Europe 4 Audi Europe 5 Audi Europe 6 Audi Europe 7 Audi Europe 8 Audi Europe 9 Audi Europe 10 Audi Europe 11 Audi Europe 12 Audi Europe 13 Audi Europe 14 BMW Europe 15 BMW Europe 16 BMW Europe 17 BMW Europe 18 BMW Europe 19 BMW Europe 20 BMW Europe 21 BMW Europe 22 BMW Europe 23 BMW Europe 24 BMW Europe 25 BMW Europe 26 BMW Europe 27 Jaguar Europe 28 Jaguar Europe 29 Jaguar Europe 30 Jaguar Europe 31 Jaguar Europe 32 Jaguar Europe 33 Jaguar Europe 34 Jaguar Europe 35 MINI Europe 36 MINI Europe 37 Mercedes-Benz Europe 38 Mercedes-Benz Europe 39 Mercedes-Benz Europe 40 Mercedes-Benz Europe 41 Mercedes-Benz Europe 42 Mercedes-Benz Europe 43 Mercedes-Benz Europe 44 Mercedes-Benz Europe 45 Mercedes-Benz Europe 46 Mercedes-Benz Europe 47 Mercedes-Benz Europe 48 Mercedes-Benz Europe 49 Mercedes-Benz Europe 50 Mercedes-Benz Europe 51 Mercedes-Benz Europe 52 Mercedes-Benz Europe 53 Saab Europe 54 Saab Europe 55 Saab Europe 56 Saab Europe 57 Saab Europe 58 Saab Europe 59 Volkswagen Europe 60 Volkswagen Europe 61 Volkswagen Europe 62 Volkswagen Europe 63 Volkswagen Europe 64 Volkswagen Europe 65 Volkswagen Europe 66 Volkswagen Europe 67 Volkswagen Europe 68 Volkswagen Europe 69 Volkswagen Europe 70 Volvo Europe 71 Volvo Europe 72 Volvo Europe 73 Volvo Europe 74 Volvo Europe 75 Volvo Europe 76 Volvo Europe 77 Volvo Europe 78 Volvo Europe What am I doing wrong? Why is the ods text printed after title2? Why is title2 the same for both tables?

peter_sjogarde · ‎04-10-2015

Hi, I have a problem importing a delimited text file into SAS. I get an error message about invalid string. I have seen that one sollution of this problem could be to change the session character encoding in SAS from wlatin1 to utf-8. I have tried to change the sasv9.cfg file line "-ENCODING wlatin1" to "-ENCODING utf-8", but this is not working. Infact nothing is working after that change, error message about connection failure to local server is given. How do I change the session character encoding? Thank you for any help!

peter_sjogarde · ‎03-20-2015

I managed to do the clustering with the SubGraphs macro. I divided the data into sections of about 1 million records and iterated over these datasets. I used a script found here: http://snipplr.com/view/6212/ The hash factor was set to 17. I do not know if that is optimal but the script was fast so that was probably a good level. Many thanks for all the help I got on this issue!

peter_sjogarde · ‎03-20-2015

Thank you for all answers helping me move forward in this process. I tried iterating over the data creating a dataset for each block and running the Subgraphs macro for each dataset. This was (of course) to slow, but it seemed to work. I will take a look at the script to see if I could add a BY-option to the SubGraphs code (any help in this matter would be greatly appreciated). Astounding – That is an acceptable result. Unfortunately I am new to hash, so I would probably not be able to putting such a script together. I realize that I need to study the hash object.

peter_sjogarde · ‎03-19-2015

I have restructured the indata so that the observations is divided into blocks, so the clustering can be done within each block which probably will faciliate the task. block id1 id2 a 1 2 a 1 3 a 4 5 a 5 2 a 6 7 b 1 2 b 1 3 ballardw - I was trying to make the data in countiguous order earlier on but I could not figure out a way to do that. It would probably be a good sollution.

peter_sjogarde · ‎03-19-2015

There are about 1.8 million unique identifiers and they are numeric. The cluster sizes will probably be from 2 to about 2000. If only connected identifiers are clustered I will not end up in one cluster, since I know that most observations are not linked. There will probably be something about 5-10 observations in each cluster in average, but I expect a skewed distribution so most clusters will only consist of a few observations.

Online Status	Offline
Date Last Visited	‎10-31-2017 08:16 AM

Re: Convergence warning when running proc quantreg

Re: Convergence warning when running proc quantreg

Re: Convergence warning when running proc quantreg

Re: Convergence warning when running proc quantreg

Re: Convergence warning when running proc quantreg

Re: Convergence warning when running proc quantreg

Convergence warning when running proc quantreg

Re: Print one title for mutiple tables within ODS

Re: Print one title for mutiple tables within ODS

Print one title for mutiple tables within ODS

Re: Convergence warning when running proc quantreg

Re: Print one title for mutiple tables within ODS

Re: Convergence warning when running proc quantreg

Re: Convergence warning when running proc quantreg

Re: Convergence warning when running proc quantreg

Re: Convergence warning when running proc quantreg

Re: Convergence warning when running proc quantreg

Re: Convergence warning when running proc quantreg

Convergence warning when running proc quantreg

Re: Print one title for mutiple tables within ODS

Re: Print one title for mutiple tables within ODS

Print one title for mutiple tables within ODS

Changing session character encoding from wlatin1 to utf-8

Re: Create clusters from pairs

Re: Create clusters from pairs

Re: Create clusters from pairs

Re: Create clusters from pairs