I have the following data coming from a stochastic experiment. I ran the experiment for 10 times for each x and tabulated the mean and standard deviations. How can I use both the mean and standard deviation to train a surrogate model? I tried using fitrgp but it does not take standard deviations. I could use sigma, but it takes a scaler value, not a vector.
x = [5; 7; 9; 11; 13];

mean_output= [103.78; 108.84; 117.68; 109.57; 72.26];
std_output = [4.20; 3.44; 4.25; 10.09; 10.71];

gprModels = fitrgp(x, mean_output, …
‘KernelFunction’, ‘squaredexponential’, …
‘BasisFunction’, ‘constant’, …
‘FitMethod’, ‘exact’, …
‘PredictMethod’, ‘exact’)I have the following data coming from a stochastic experiment. I ran the experiment for 10 times for each x and tabulated the mean and standard deviations. How can I use both the mean and standard deviation to train a surrogate model? I tried using fitrgp but it does not take standard deviations. I could use sigma, but it takes a scaler value, not a vector.
x = [5; 7; 9; 11; 13];

mean_output= [103.78; 108.84; 117.68; 109.57; 72.26];
std_output = [4.20; 3.44; 4.25; 10.09; 10.71];

gprModels = fitrgp(x, mean_output, …
‘KernelFunction’, ‘squaredexponential’, …
‘BasisFunction’, ‘constant’, …
‘FitMethod’, ‘exact’, …
‘PredictMethod’, ‘exact’) I have the following data coming from a stochastic experiment. I ran the experiment for 10 times for each x and tabulated the mean and standard deviations. How can I use both the mean and standard deviation to train a surrogate model? I tried using fitrgp but it does not take standard deviations. I could use sigma, but it takes a scaler value, not a vector.
x = [5; 7; 9; 11; 13];

mean_output= [103.78; 108.84; 117.68; 109.57; 72.26];
std_output = [4.20; 3.44; 4.25; 10.09; 10.71];

gprModels = fitrgp(x, mean_output, …
‘KernelFunction’, ‘squaredexponential’, …
‘BasisFunction’, ‘constant’, …
‘FitMethod’, ‘exact’, …
‘PredictMethod’, ‘exact’) surrogate, fitrgp MATLAB Answers — New Questions

​