simulation multivariate normal data with 10 variables

thanoon — Thu, 12 Dec 2013 02:56:12 GMT

hi all

i need your help to simulate multivariate normal data with 10 var. when i run this code i receive errors .

%let N = 1000; /* size of each sample */

/* Multivariate normal data */

proc iml;

/* specify the mean and covariance of the population */

Mean = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10};

Cov = {32.0897 13.1740 -4.8130 -2.9651 6.0853 2.2018 2.6580 3.4379 -1.8501 -7.4523

13.1740 18.3444 -5.2602 -9.4638 3.0731 2.1855 3.9010 3.0226 -3.5204 -4.3176

-4.8130 -5.2602 7.7173 2.1597 -0.3544 -3.3951 -0.7576 -2.9107 5.3209 5.5138

-2.9651 -9.4638 2.1597 20.1232 1.1670 0.8156 -12.3116 -2.4223 5.5468 2.1305

6.0853 3.0731 -0.3544 1.1670 3.8644 0.6464 0.5127 0.9321 -0.3289 -1.9678

2.2018 2.1855 -3.3951 0.8156 0.6464 7.8445 2.9060 -1.2132 -4.2227 -5.0253

2.6580 3.9010 -0.7576 -12.3116 0.5127 2.9060 14.5232 1.7516 -4.8958 1.2902

3.4379 3.0226 -2.9107 -2.4223 0.9321 -1.2132 1.7516 6.6480 -0.7609 2.7944

-1.8501 -3.5204 5.3209 5.5468 -0.3289 -4.2227 -4.8958 -0.7609 7.9543 5.7160

-7.4523 -4.3176 5.5138 2.1305 -1.9678 -5.0253 1.2902 2.7944 5.7160 16.1720};

call randseed(4321);

X = RandNormal(&N, Mean, Cov); /* 1000 x 10 matrix */

/* check the sample mean and sample covariance */

SampleMean = mean(X); /* mean of each column */

SampleCov = cov(X); /* sample covariance */

/* print results */

c = "x1":"x10";

print (X[1:,1000]);

print SampleMean[colname=c];

print SampleCov[colname=c rowname=c];

/* write SAS/IML matrix to SAS data set for plotting */

create MVN from X[colname=c]; append from X; close MVN;

quit;

Re: simulation multivariate normal data with 10 variables

thanoon — Thu, 12 Dec 2013 08:20:44 GMT

hi all

i solved this problem, i forget the comma for covariance matrix.

regards