If it is indeed the case that every observation has at least one missing value, you might need to choose another approach. Three ideas: 1) There might be one or two (or 10) variables that contain mostly missing values, whereas the other 999 variables don't contain ANY missings. If so, you can exclude those "bad" variables and just use the good ones. To count the number of missing values for each variable, use proc means data=IN NMISS N; var VAR01-Var200; run; 2) If your missing values are scattered at random throughout the variables, you might be able to impute values for the missings by using PROC MI, and then analyze the imputed variables.  However, if you have too many missing values (like 60%). I'm not sure I would trust the MI results. 3) Instead of VARCLUS, you can use PROC CORR to look at PAIRWISE correlations between variables.  A pairwise approach uses all observations in common between a pair of variables, whereas the LISTWISE approach that you are currently using requires all 200 variables be nonmissinging.There are then various ways to cluster variables based on the pairwise correlations. For example, see C. Hurley (2004) "Clustering Visualisations of Multidimensional Data", J. Comp. and Graphical Stats, 13(4), pp 788-806 ... View more

I hope you have SAS 9.2 or better. You can use the ZIPCITYDISTANCE function to compute the distances: http://support.sas.com/kb/36/091.html It shouldn't take too long: for each firm, find the minimum of the distances to the 50 citites. You can do this in a DATA step. ... View more

I'm not an expert, but it sounds like you might want to look up McNemar's test, which is one of the tests for agreement. See the documentation at http://support.sas.com/documentation/cdl/en/statug/63962/HTML/default/viewer.htm#statug_freq_a0000000568.htm For an example, see http://www.ats.ucla.edu/stat/sas/whatstat/whatstat.htm (search for "McNemar's Test" on the page) data a; input survey q freq; datalines; 0 0 27 0 1 22 1 0 17 1 1 23 ; proc freq data=a; weight freq; tables survey*q / agree; /* or EXACT mcnem; for small samples */ run; ... View more

For variables that encompass several order of magnitudes, a good rule of thumb is to apply a log transformation. For these data, I'd try y=log10(purchases_amount) and use Y as the response variable. Also, PROC REG will compute regression diagnostic plots for you. Just turn on ODS graphics: ods graphics on; The "residual vs. predicted" [;pt is the upper left in a panel of diagnostic plots. See Figure 76.12 in the PROC REG doc: http://support.sas.com/documentation/cdl/en/statug/63962/HTML/default/viewer.htm#statug_reg_sect004.htm#statug.reg.gsEx2PredictionPanel ... View more

To Authur's reply, I'll add that the expression date = mdy(02,29,year); returns a valid value is YEAR is a leap year and a missing value otherwise. ... View more

Jackson (1991), A user's guide to principal components, is an excellent book on PCA. There is a section on PCA vs linear discriminant analysis (LDA) on pp. 334-337, including references to the use of PCA in discriminant analysis. A more recent book is Hastie, Tibshirani, and Friedman, The Elements of Statistical Learning, which is more advanced, but less explicit than Jackson. The issue raised in the web page you quote is covered on pp. 93-94. In general, PCA and LDA are two different analyses with different objectives. PCA tries to find directions (=linear combinations fo variables) that explain the most variance. LDA tries to find a linear subspace that best separates groups.  Tjhey can be combined into canonical discriminant analysis, which in SAS is accomplished by using the CANDISC procedure. The overview section of the CANDISC doc discusses how CDA works. From a practical perspective, CDA works pretty well in many cases. ... View more

Questions at all levels are welcome. It is often difficult for a beginner to get started, especially when you are coming from another language like MATLAB. TomTom correct mentions that there are some resources that might help you with the learning curve: - the SAS/IML Tip sheet - a short (and not complete) "MATLAB to SAS/IML Tip Sheet" - the FREE chapter of my book [HIGHLY RECOMMENDED] - Read the article "How to learn SAS/IML: Five resources for the beginner" In the "How to learn SAS/IML" article, the 3rd suggestion is "ask questions at the SAS/IML discussion forum"! So (IMHO), feel free to ask for help when you get stuck. ... View more

The matlab syntax is NSTART: INCR: NSTOP. You can do the same in SAS/IML by using the DO function: x = do( nstart, nstop, incr); For example, x = do( 1, 45, 4); y = do( 2, 45, 4); For more on this topic (including the SAS/IML equivalent of the linspace() function in MATLAB), see http://blogs.sas.com/content/iml/2011/01/10/creating-vectors-that-contain-evenly-spaced-values/ If you need to generate points on a 2D grid, see http://blogs.sas.com/content/iml/2011/01/21/how-to-create-a-grid-of-values/ ... View more

You can't list multiple matrices on the RETURN statement (that's MATLAB syntax) but you can return as many values as you want by writing a subroutine module instead of a function module. For an example, see the bottom of p. 2 of http://support.sas.com/resources/papers/proceedings10/329-2010.pdf Incidentally, you can do this because SAS/IML passes arguments by reference, not by value. The SAS/IML convention is that output arguments are listed first, so your module might look like proc iml; start ols (beta, residual, y,x);   beta=inv(x`*x)*(x`*y);   residual=y-x*beta; finish ols; /* define data */ x= { 1 1 1,     1 2 4,     1 3 9,     1 4 16,     1 5 25,     1 6 36,     1 7 49,     1 8 64 }; y= {3.929,5.308,7.239,9.638,12.866,17.069,23.191,31.443}; /* call module */ run ols(b, r, y, x); print b, r; ... View more

I'm not sure what you mean by an axis legend. Do you mean that you want the variables names (or labels) to be displayed next to the axes?  If so, I don't think you can get that with the MATRIX statement, but you can use the PLOT or COMPARE statements. For example,  the following code arranges two plots in a panel and labels all axes: PROC SGSCATTER  DATA=sashelp.class; compare x=(height age) y=weight / GROUP=sex; RUN; ... View more

If the X axis is reversed, this looks like a classic case of heteroskedastic data in which the variance increases with X. I'd define Z = 1/X and plot Y vs. Z (or use 1/(X+1) if X=0 is a possible value). You can then use all the standard variance stabilizing transformations (AKA, normalizing transformations) such as Z-->log(Z) or  Z-->sqrt(Z). Depending on the meaning of X, you could also try just a simple reflection such as W = 300000 - x and then work with W. As Steve mentions, transformations often work best when they are meaningfully related to the data. So ask yourself, is "natural" way to flip around the X axis? ... View more