hi
I have made a model in which a surgeon performs surgeries on multiple patients. The surgeon is capable of performing 2 different types of surgeries (type A and type B). type A surgery is more desirable but at the same time, it requires more skills from the surgeon compared to type B surgery.
It is assumed that the surgeon of our model has a fixed level of skills but the required skill to perform type A surgery on different patients varies depending on their demographic and clinical characteristics.
In order to simulate the different level of difficulty of performing type A surgery in different patients, a random number is assigned to each patient showing the level of skills required from the surgeon to enable him to perform type A surgery on that patient.
clearly, if the skill of the surgeon is equal to or more than the skill required to perform surgery on a patient (random number), type A surgery will be performed otherwise, type B is performed.
I have used “Pink Noise” with the following formulation to generate random number in this model, is this way of generating random number correct in this problem? of not, what is the right way of generating random numbers in this problem.
thanks in advance
noise seed = 1
unit = Dmnl
mean = 0
unit = skill
standard deviation = 1
unit = skill
time step = 0.015625
unit = year
correlation time = 2
unit = year
white noise = RANDOM NORMAL( -1e+06, 1e+06, mean, standard deviation*((2-TIME STEP/correlation time)/(TIME STEP/correlation time))^0.5 , noise seed)
Unit = skill
change in pink noise = (white noise-pink noise)/correlation time
unit = skills/Year
pink noise = INTEG (change in pink noise)
initial value = RANDOM NORMAL(-100, 100, mean, standard deviation, noise seed)
unit = skill
what method of generating random variable is correct in this problem?
Re: what method of generating random variable is correct in this problem?
This particular subforum doesn't get many eyeballs, I think.
I don't see a strong reason for pink noise here, because there's no obvious way that patients' characteristics would be correlated over time. Seems more like a case for white noise (i.e. RANDOM NORMAL or similar).
I don't see a strong reason for pink noise here, because there's no obvious way that patients' characteristics would be correlated over time. Seems more like a case for white noise (i.e. RANDOM NORMAL or similar).
/*
Advice to posters (it really helps us to help you)
http://www.ventanasystems.co.uk/forum/v ... f=2&t=4391
Blog: http://blog.metasd.com
Model library: http://models.metasd.com
Bookmarks: http://delicious.com/tomfid/SystemDynamics
*/
Advice to posters (it really helps us to help you)
http://www.ventanasystems.co.uk/forum/v ... f=2&t=4391
Blog: http://blog.metasd.com
Model library: http://models.metasd.com
Bookmarks: http://delicious.com/tomfid/SystemDynamics
*/