Samples & SAS Notes > Analytics
>
Descriptive Statistics
| Date | Title | Type | ||
|---|---|---|---|---|
| 11/20/09 |
37767 - How can I produce a box plot with a logarithmically-scaled vertical axis? PROC BOXPLOT does not support nonlinear axis scaling, so it has no PLOT statement options for a log scale. Also, scaling options on an AXIS statement are ignored. Beginning in SAS 9.2, you can generate a box plot with ... http://support.sas.com/kb/37/767.html, 24KB |
Usage Note | ||
| 10/26/09 |
24822 - Changing the Numeric Format of PROC FREQ Output This sample describes how to change these defaults for the three main types of PROC FREQ output: one-way tables, multiway tables in list format, and multiway tables presented as crosstabulations. http://support.sas.com/kb/24/822.html, 21KB |
Sample | ||
| 8/28/09 |
37000 - How to produce a scatter plot matrix of ranked variables if SPEARMAN correlations are requested When Spearman correlations are requested in PROC CORR by specifying the SPEARMAN option, the correlations are computed by ranking the data and using the ranks in the Pearson product-moment correlation formula. However, ... http://support.sas.com/kb/37/000.html, 44KB |
Usage Note | ||
| 8/21/09 |
36453 - How to import and assess outside modeling techniques using SASŪ Enterprise Miner(tm) This video shows an example of how to import a non-supplied model into SAS Enterprise Miner and be able to assess it with other models. http://support.sas.com/kb/36/453.html, 21KB |
Usage Note | ||
| 7/7/09 |
36448 - IML Studio may fail to start when IML is not licensed IML Studio may fail to start even if SAS is installed locally when IML is not licensed http://support.sas.com/kb/36/448.html, 19KB |
Problem Note | ||
| 6/24/09 |
35900 - Only the first 5 analysis variables are graphed in the scatter plot matrix By default, the ODS Graphics scatter plot matrix generated with the PLOTS=MATRIX option displays 5 variables. You can control the number of variables displayed with the NVAR= matrix option. Note this option is named ... http://support.sas.com/kb/35/900.html, 19KB |
Usage Note | ||
| 4/29/09 |
14469 - Variables reporting N in the ODS OUTPUT data sets are not present if no variable used to calculate the correlations contains a missing value If no analysis variable contains a missing value and if no BY variables are used, the variables that report N in the PEARSONCORR, SPEARMANCORR, HOEFFDINGCORR, and KENDALLCORR ODS OUTPUT data sets will not be present. ... http://support.sas.com/kb/14/469.html, 22KB |
Problem Note | ||
| 4/28/09 |
18827 - The Two Sample Test for Proportions is incorrect in the Analyst Application The Two Sample Test for Proportions in the Analyst Application will generate incorrect results if the cell count for any cell is zero. To circumvent the problem, use the CHISQ option in the TABLE statement in PROC FREQ ... http://support.sas.com/kb/18/827.html, 19KB |
Problem Note | ||
| 2/26/09 |
22991 - How can I remove the title "The FREQ Procedure" from my FREQ procedure output? You can remove "The FREQ Procedure" title from your FREQ procedure output by inserting the ODS NOPROCTITLE statement before the PROC FREQ statement. ODS NOPROCTITLE; PROC FREQ; TABLE X*Y; RUN; After you have ... http://support.sas.com/kb/22/991.html, 17KB |
Usage Note | ||
| 2/23/09 |
34843 - Incorrect schematic box-and-whisker plots with greater than 32,768 outliers If the BOXSTYLE=SCHEMATIC, SCHEMATICID or SCHEMATICIDFAR option is specified on the PLOT statement in PROC BOXPLOT and the number of outliers beyond the upper or lower fence or far fence is greater than 32,768, then the ... http://support.sas.com/kb/34/843.html, 27KB |
Problem Note | ||
