https://github.com/beliavsky/multivariate-normal-random-deviates
Generate multivariate normal random deviates in Fortran
https://github.com/beliavsky/multivariate-normal-random-deviates
multivariate-normal-distribution normal-distribution random rng simulation statistics
Last synced: 2 months ago
JSON representation
Generate multivariate normal random deviates in Fortran
- Host: GitHub
- URL: https://github.com/beliavsky/multivariate-normal-random-deviates
- Owner: Beliavsky
- License: mit
- Created: 2025-02-21T02:13:20.000Z (3 months ago)
- Default Branch: main
- Last Pushed: 2025-02-21T02:38:33.000Z (3 months ago)
- Last Synced: 2025-02-21T03:26:37.924Z (3 months ago)
- Topics: multivariate-normal-distribution, normal-distribution, random, rng, simulation, statistics
- Language: Fortran
- Homepage:
- Size: 11.7 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Multivariate Normal Random Deviates
Multiplies uncorrelated normal deviates by a Cholesky factor of the covariance matrix to generate
multivariate normal deviates with a specified covariance matrix. Sample output:
```
#obs: 1000000
true means and standard deviations
Variable 1: Mean = -5.00000000, StdDev = 10.00000000
Variable 2: Mean = 0.00000000, StdDev = 20.00000000
Variable 3: Mean = 5.00000000, StdDev = 30.00000000
empirical means and standard deviations
Variable 1: Mean = -4.99696615, StdDev = 9.99908359
Variable 2: Mean = 0.00859363, StdDev = 19.99599058
Variable 3: Mean = 5.00536297, StdDev = 30.01204086
true correlation matrix:
1.000000 0.500000 0.300000
0.500000 1.000000 0.400000
0.300000 0.400000 1.000000
empirical correlation matrix:
1.000000 0.499548 0.299604
0.499548 1.000000 0.400492
0.299604 0.400492 1.000000
maxval(abs(xcorr - emp_corr)): 0.000492
empirical means and standard deviations
Variable 1: Mean = -0.01287349, StdDev = 9.99949712
Variable 2: Mean = 0.00482072, StdDev = 20.00958772
Variable 3: Mean = -0.02900624, StdDev = 30.02427445
empirical correlation matrix:
1.000000 0.500036 0.299948
0.500036 1.000000 0.400134
0.299948 0.400134 1.000000
maxval(abs(xcorr - emp_corr)): 0.000134
```