07-Jan-2022 18:14:48 correlation_test(): MATLAB/Octave version 9.8.0.1380330 (R2020a) Update 2. Test CORRELATION. CORRELATION_TEST01 Make plots of correlation functions. Created "besselj_plot.png". Created "besselk_plot.png". Created "correlation_circular_plot.png". Created "constant_plot.png". Created "cubic_plot.png". Created "damped_cosine_plot.png". Created "damped_sine_plot.png". Created "exponential_plot.png". Created "gaussian_plot.png". Created "hole_plot.png". Created "linear_plot.png". Created "matern_plot.png". Created "pentaspherical_plot.png". Created "power_plot.png". Created "rational_quadratic_plot.png". Created "spherical_plot.png". Created "correlation_white_noise_plot.png". CORRELATION_TEST02 SAMPLE_PATHS_CHOLESKY generates sample paths from the correlation matrix, factored using the Cholesky factor. It requires that the correlation matrix is nonnegative definite. SAMPLE_PATHS_EIGEN generates sample paths from the correlation matrix, factored using the eigen factorization. If the correlation matrix is not nonnegative definite, we simply suppress negative eigenvalues. SAMPLE_PATHS_FFT generates sample paths from the correlation matrix, factored using the FFT factorization of the correlation matrix after embedding in a circulant. Created "besselj_paths.png". Created "besselk_paths.png". Created "circular_paths.png". Created "constant_paths.png". Created "cubic_paths.png". Created "damped_cosine_paths.png". Created "damped_sine_paths.png". Created "exponential_paths.png". Created "gaussian_paths.png". Created "hole_paths.png". Created "linear_paths.png". Created "matern_paths.png". Created "pentaspherical_paths.png". Created "power_paths.png". Created "rational_quadratic_paths.png". Created "spherical_paths.png". Created "white_noise_paths.png". CORRELATION_TEST03 Make plots of correlation functions. Graphics saved as "besselj_plots.png". Graphics saved as "besselk_plots.png". Graphics saved as "circular_plots.png". Graphics saved as "constant_plots.png". Graphics saved as "cubic_plots.png". Created "damped_cosine_plots.png". Graphics saved as "damped_sine_plots.png". Graphics saved as "exponential_plots.png". Graphics saved as "gaussian_plots.png". Graphics saved as "hole_plots.png". Graphics saved as "linear_plots.png". Graphics saved as "matern_plots.png". Graphics saved as "pentaspherical_plots.png". Graphics saved as "power_plots.png". Graphics saved as "rational_quadratic_plots.png". Graphics saved as "spherical_plots.png". Graphics saved as "white_noise_plots.png". CORRELATION_TEST04 Convert between a correlation and a covariance matrix. Covariance matrix K: Col: 1 2 3 4 5 Row 1 : 1 1 1 1 1 2 : 1 2 2 2 2 3 : 1 2 3 3 3 4 : 1 2 3 4 4 5 : 1 2 3 4 5 Correlation matrix C: Col: 1 2 3 4 5 Row 1 : 1 0.707107 0.57735 0.5 0.447214 2 : 0.707107 1 0.816497 0.707107 0.632456 3 : 0.57735 0.816497 1 0.866025 0.774597 4 : 0.5 0.707107 0.866025 1 0.894427 5 : 0.447214 0.632456 0.774597 0.894427 1 Variances: 1: 1 2: 1.41421 3: 1.73205 4: 2 5: 2.23607 Recovered covariance matrix K2: Col: 1 2 3 4 5 Row 1 : 1 1 1 1 1 2 : 1 2 2 2 2 3 : 1 2 3 3 3 4 : 1 2 3 4 4 5 : 1 2 3 4 5 CORRELATION_TEST05 CORRELATION_BROWNIAN_DISPLAY displays the Brownian correlation Created "correlation_brownian_plots.png". CORRELATION_TEST06 For non-stationary correlation functions: SAMPLE_PATHS2_CHOLESKY generates sample paths from the correlation matrix, factored using the Cholesky factor. It requires that the correlation matrix is nonnegative definite. SAMPLE_PATHS2_EIGEN generates sample paths from the correlation matrix, factored using the eigen factorization. If the correlation matrix is not nonnegative definite, we simply suppress negative eigenvalues. SAMPLE_PATHS2_FFT generates sample paths from the correlation matrix, factored using the FFT factorization of the correlation matrix after embedding in a circulant. Created "correlation_brownian_paths.png". CORRELATION_TEST Normal end of execution. 07-Jan-2022 18:15:04