MATLAB Need help with Matlab standard deviation

AI Thread Summary
The discussion revolves around coding a vaccination simulation in MATLAB to measure infection rates over 100 days, focusing on the variable delta, which represents the infection period. The average delta is 5 days, but it varies among individuals. The user seeks to generate a random standard deviation for a log-normal distribution to use in the formula R = lognrd(5, ?, 100, 10), where 5 is the mean. A key point is that there is no fixed standard deviation for a log-normal distribution; it must be defined as an input. The conversation highlights the need to clarify the differences between normal and log-normal distributions, with an emphasis on the flexibility of the standard deviation in shaping the distribution of infection periods.
spoonyluv
Messages
9
Reaction score
0
Hello,

I seriously need some help as I can't figure out what to do here. I am working on coding a vaccination simulation that measures rates of infection for period of 100 days.

I have this variable delta, which is the number of days the infection period lasts. Delta is 5, or at least on average it is 5, which means it could be occasionally be 1 day for a random individual or 3 days for some other random individual or even 7, but on average it is 5.

I have 10 individuals and run the simulation for 100 days. for each individual i need MATLAB to create a random standard deviation that is part of a normal distribution. So for example if I have a lognormal formula

R = lognrd (5, ?, 100, 10)

This formula creates a perfect matrix of 100x10 with lognormal distribution values for each cell, with 5 being the mean. The problem is with the ?, which is where the standard deviation goes. How do I get MATLAB to churn out a std dev of a normal distribution that will then feed into the R = lognrd (5, ?, 100, 10)

Thanks
 
Physics news on Phys.org
There is no "standard" standard deviation for a log-normal distribution. See: http://en.wikipedia.org/wiki/Log-normal_distribution. This is another input that you need to give to MATLAB, not the other way around.

I other words, you can have a distribution where 99% of the population has an infection period of 4-6 days, with the other 1% being "other." You could also have a distribution where 50% of the population has an infection period of 4-6 days, with the other half being "other." In both cases it is possible for the distribution to be log-normal.

-Kerry
 
Oooh, I think I'm wrong in the above post (or at least I'm not sure I'm right). I responded because I've been looking at normal distributions recently and they were fresh in my mind - but these are NOT the same as a log-normal distribution.

My advice - ignore the above post :-)

Sorry for the confusion...

-Kerry
 

Similar threads

Replies
2
Views
1K
Replies
5
Views
3K
Replies
2
Views
2K
Replies
4
Views
1K
Replies
24
Views
5K
Replies
1
Views
2K
Replies
2
Views
2K
Replies
10
Views
2K
Back
Top