Generating Normally Distributed Integers in R, Matlab, and Other Software

AI Thread Summary
Generating normally distributed integers in R or Matlab poses challenges, as a distribution limited to integers cannot be normal. The discussion highlights the need for normally distributed inter-arrival times, such as those represented in time intervals. Rounding generated decimal values may compromise the normality of the data, and using a normal distribution could lead to negative values, which are not feasible for arrival times. It is suggested to consider alternative distributions, like exponential, which naturally accommodate positive values. Understanding the implications of the chosen distribution is crucial for accurate modeling of arrival times.
Mark J.
Messages
81
Reaction score
0
Hi.
Any ideas how to generate in any software R, Matlab etc normally distributed random numbers by one condition so they will be integers like 1, 2,3...and not decimal values.
I tried to round up the generated decimal values but in my guess normality is doubted after that.
Regards
 
Physics news on Phys.org
You will have to be clearer than this. A distribution which only takes on integers can never be normal.

What is it you really want??
 
Thanks,
I mean I need to generate some sample data like 9.00 AM 9.04 AM 9.07 AM but with condition that inter-arrival times are normally distributed.
That's why I need some numbers like 4 (9.04AM -9.00AM ) etc normally distributed.
Any help on this?
 
Arrivals are continuous. Your numbers will be real. You can round them if you like but if you do that then the approximation to normal depends on the parameters of the distribution.

I don't know what you're trying to do, but has it occurred to you that modeling arrivals with a normal distribution allows for the possibility of an arrival occurring before the previous arrival? You might want to consider that.
 
Can you please explain something more about?
Regards
 
I don't know which part you're referring to. For the first, just consider a normal density with s.d.=.01 vs. one with a very large s.d. If you draw from the first and round everything to the nearest integer then you basically end up with the mean for virtually all of your points. In the second case you could round to the nearest integer and still "approximate" normal, whatever that means in your context.

As to the second point, inter-arrival times are positive. T1 is the time from zero to the first arrival, T2 is the time from T1 of the second arrival, etc. So the times are usually described with a density that is zero for t<0 such as an exponential. If you use a normal distribution then you permit negative values and risk that, for example, T2<0, and your second arrival occurred prior to your first. You would take this into account in the context of what exactly you are trying to do.
 
I was reading documentation about the soundness and completeness of logic formal systems. Consider the following $$\vdash_S \phi$$ where ##S## is the proof-system making part the formal system and ##\phi## is a wff (well formed formula) of the formal language. Note the blank on left of the turnstile symbol ##\vdash_S##, as far as I can tell it actually represents the empty set. So what does it mean ? I guess it actually means ##\phi## is a theorem of the formal system, i.e. there is a...
Back
Top