About fri0

fri0 · ‎07-27-2016

Hi ryall! thanks for your answers. I used the default options in Cluster node, the result it's a ward clustering with abouth 10 clusters and also it has a giant cluster... I used the CCC plot to identify a possible optimal number of cllusters, with the graph I identified between 4 or 5 clusters. That's why I used a 4 and 5 k-means clustering, but the result it's almost the same with one cluster taking almost all the registers. I only have 2 intervals variables: total amount and number of operations. Any other idea? I found that sometimes use replacement and filter node could help.

fri0 · ‎07-26-2016

Hi, I've run a k-means clustering in Enterprise Miner, but I get a giant cluster with 99% of the registers.... Any idea to get clusters more equitable? Thanks!

fri0 · ‎05-04-2016

I can't find any option in clustering node to save the centroids, the output save only cluster and distance. Do you have some pdf, tutorial or link about this?

fri0 · ‎05-04-2016

Hi, I've created a k-means model using SAS Enterprise Guide, but I don't know how to use it to score/clasify another dataset. I'd like to know how to extract the code of the model to use in another query if it's possible. Thanks in advance.

fri0 · ‎04-21-2016

Thans PG, this work absolutely perfect.

fri0 · ‎04-21-2016

Thanks RW9, this work perfect when there is no missing space as first value for last. But, there were cases with first value missing, so I had to chose the PGSTats.

fri0 · ‎04-19-2016

The SAS server is down, but I'll try when it'll be operative again.

fri0 · ‎04-19-2016

Hi, I need your help to complete a dataset. I have many stock shares ordered by share and date. Also I have its last price of the day, but sometimes there is a missing value if there was no trading that day. I want to complete missing values with the last valid value. The amount of missing values are different by each stock shares. Thanks in advance for your help. This is an example of my dataset: Name Date Last AAA 4/1/16 5.10 AAA 5/1/16 . AAA 6/1/16 . AAA 7/1/16 . AAA 8/1/16 5.13 AAA 11/1/16 5.13 AAA 12/1/16 5.15 BBB 4/1/16 25.60 BBB 5/1/16 25.61 BBB 6/1/16 . BBB 7/1/16 . BBB 8/1/16 25.65 BBB 11/1/16 . BBB 12/1/16 25.66 I need this: For the second register, last price is missing, so I need to replace it with the previous value 5.10. The third register also has no value for last, so I replace with the previous value 5.10 and then until find a valid value. Name Date Last AAA 4/1/16 5.10 AAA 5/1/16 5.10 AAA 6/1/16 5.10 AAA 7/1/16 5.10 AAA 8/1/16 5.13 AAA 11/1/16 5.13 AAA 12/1/16 5.15 BBB 4/1/16 25.60 BBB 5/1/16 25.61 BBB 6/1/16 25.61 BBB 7/1/16 25.61 BBB 8/1/16 25.65 BBB 11/1/16 25.65 BBB 12/1/16 25.66

fri0 · ‎10-22-2015

Thanks PG. I couldn't select your answer as solution. It seems you could choose only one.

fri0 · ‎10-22-2015

Pg, thanks, I tried your solution and it works perfectly!

fri0 · ‎10-22-2015

My bad, I just see my error, id is a parameter, I was confused because I have the same name for my variable...it works perfect!

fri0 · ‎10-22-2015

Hi PG, a friend told me that proc sql is a method that take more time, is that truth? What would you choose between proc summary and proc sql?? Because I have about 500 million obs.

fri0 · ‎10-22-2015

Hi Jerry, thanks for your answer, but there is not totally right: proc summary output: id date sum_mto 0 OCT14 0.2611745 0 NOV14 0.01 0 DEC14 0.08014064 0 FEB15 0.47 0 MAR15 0.12968991 0 APR15 0.37 0 MAY15 0.02316756 0 JUN15 0.05 0 JUL15 0.05017248 0 AUG15 0.26142127 0 SEP15 0.01620732 2 APR15 15.044794 2 JUN15 0.0317058 83 DEC14 300001.67 262 SEP15 31.0078 310 NOV14 3257 361 JUN15 15000 382 OCT14 20.5578741 526 MAR15 0.8397922 583 DEC14 0.00669792 584 SEP15 2.96077683 603 APR15 35.11013 603 JUN15 69.369564 609 OCT14 68.6342 693 JAN15 5.609508 693 JUN15 25.26848 710 JUN15 1734.04944 718 JUL15 0.00313578 727 MAY15 0.00316756 742 JAN15 23.25582 793 MAY15 0.00633512 881 SEP15 80.5151477 921 AUG15 12.31 935 MAY15 63.7552 948 DEC14 353.77335 948 MAR15 18.398973 I added var to your code, is it right? BUt, here is the problem, the output is strange: proc transpose data=summ out=want(drop=_: ) prefix=mto_ ; by id; var id date; run; The output of this code: id mto_1 mto_2 mto_3 mto_4 mto_5 mto_6 mto_7 mto_8 mto_9 mto_10 mto_11 mto_12 0 0 0 0 0 0 0 0 0 0 0 0 . 0 19997 20028 20058 20121 20148 20179 20209 20240 20270 20301 20332 . 2 2 2 . . . . . . . . . . 2 20179 20240 . . . . . . . . . . 83 83 . . . . . . . . . . . 83 20058 . . . . . . . . . . . 262 262 . . . . . . . . . . . 262 20332 . . . . . . . . . . . 310 310 . . . . . . . . . . . 310 20028 . . . . . . . . . . . 361 361 . . . . . . . . . . . 361 20240 . . . . . . . . . . . 382 382 . . . . . . . . . . . 382 19997 . . . . . . . . . . . 526 526 . . . . . . . . . . . 526 20148 . . . . . . . . . . . 583 583 . . . . . . . . . . . 583 20058 . . . . . . . . . . . 584 584 . . . . . . . . . . . 584 20332 . . . . . . . . . . . The date was transformed from a ddmmyy format, could it be the problem? And, you know why there are replicated ids? Thanks!

fri0 · ‎10-21-2015

Hi, I need a particular kind of summary, I hope that it would be possible to do: I have this datatable: ID MTO DATE 1 100 jul-15 2 120 jun-15 3 130 may-15 1 90 may-15 2 80 jun-15 3 100 may-15 1 200 may-15 2 110 jul-15 3 100 jul-15 I need to get kind of summary, but using data step cause I have too many obs: ID Sum(MTO) may-15 Sum(MTO) jun-15 Sum(MTO) jul-15 1 330 80 200 2 190 120 110 I was using proc summary, but I could only get a sum without the group by date.

fri0 · ‎08-17-2015

Hi, thank your very much for your answers. I have another problem. I have been questioned about the cluster proximity plot. I have only 6 cluster, but three of them are very close, almost overlapped. Further more, you could see an asterisk with a circle for each cluster, but in my graph, you could only see the asterisk. I really need help ton understand that graphic and could interpret it. Thanks in advance and sorry for my english.

Online Status	Offline
Date Last Visited	‎04-03-2019 08:45 PM

Re: Error trying to make a left join

Error trying to make a left join

Re: Problem with legend in scatterplot

Re: Problem with legend in scatterplot

Problem with legend in scatterplot

Re: Macro doesn't read the conditional

Re: Macro doesn't read the conditional

Macro doesn't read the conditional

Re: Loop with conditional range

Re: Loop with conditional range

Re: Macro doesn't read the conditional

Re: Macro doesn't read the conditional

Re: Macro doesn't read the conditional

Re: Macro doesn't read the conditional

Re: Loop with conditional range

Re: Improve K-Means results

Improve K-Means results

Re: How to use a model for another dataset in SAS Guide

How to use a model for another dataset in SAS Guide

Re: How to complete missing values with previous value

Re: How to complete missing values with previous value

Re: How to complete missing values with previous value

How to complete missing values with previous value

Re: Summary by variable

Re: Summary by variable

Re: Summary by variable

Re: Summary by variable

Re: Summary by variable

Summary by variable

Re: How could I validate clusters from k-means