Generate array with 3 gaussian distributions MATLAB

Question

I need help! My purpose is to develop in MATLAB a routine that, starting from a series of actions (modeled by a label, a mean value and a variance), is able to generate an array of activity. I explain better with my code:

action_awake_in_bed = [1 5*60 1*60];
action_out_of_bed = [3 30 10];
action_out_bedroom = [2 2*60 15];
ACTIVITY_WAKE = {'action_awake_in_bed','action_out_of_bed','action_out_bedroom'};

The first element of action array is a label (a posture label), the second element is the length of the action (in seconds), the third element the variance.

I need as output the array ACTIVITY_WAKE....

Thanks

It is not clear: are you expecting an array with labels [1 2 3] with the amount of each label is distributed with a different mean and variance? — Shai
– Shai, Commented Dec 22, 2014 at 10:53
Yes, I expect an array in output with a sequence of labels ordered by the action and distributed with the different mean and variance that I have indicate as parameter in the array of actions..For example the array "ACTIVITY_WAKE' could start with 1111111111 (for 300 time +/-60), after 3333333 (30 times +/-10) and finally 2222222 (120 times +-15) — Andrea Caroppo
– Andrea Caroppo, Commented Dec 22, 2014 at 10:59

Shai · Accepted Answer · 2014-12-22 14:52:08Z

1

Let's use a struct to store the meta-parameters

action.awake_in_bed   = [1 5*60 1*60];
action.out_of_bad     = [3 30 10];
action.out_of_bedroom = [2 2*60 15];
ACTIVITY = {'awake_in_bed','out_of_bad','out_of_bedroom'};

After these pre-definitions, we can sample an activity vector

ACTIVITY_WAKE = cell(1,numel(ACTIVITY)); 
for ii = 1:numel( ACTIVITY ) %// foreach activity
    cp = action.(ACTIVITY{ii}); %// get parameters of current activity
    n = round( cp(2) + sqrt(cp(3))*randn() ); %// get the number of samples
    ACTIVITY_WAKE{ii} = repmat( cp(1), 1, n );
end
ACTIVITY_WAKE = [ ACTIVITY_WAKE{:} ];

To get the number of samples I use the following recipe to sample from a normal distribution with mean~=0 and std~=1.

edited Dec 22, 2014 at 14:52

answered Dec 22, 2014 at 12:43

Shai

115k39 gold badges259 silver badges398 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

Andrea Caroppo Over a year ago

Are you sure about the formula to obtain n? Because I obtain always values near the mean value, with little variance...

Collectives™ on Stack Overflow

Generate array with 3 gaussian distributions MATLAB

1 Answer 1

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related