comaprison

[dw7672]

I am doing some random samples real mearsurmenst from an indoor enviornment and I am using a software to simualte this enviornment to have the readings also , I want to see the effiecny of the software , what statistical paramterers of random variable are needed to comprate two results

 

[dw7672]

no answer

 

[GregLocock]

You haven't described what you are trying to do well enough for us to help. Can you give the data you want to test?


Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[corus]

If the variables follow a normal distribution then compare the means and standard deviation with your software

corus

 

[dw7672]

it is supposed to follow a rice or rayleigh distribution

 

[m777182]

First find out what distribution do the samples belong to most likely. Then make statistical moments (first,second,...). Then you can find a possible trend, if there is some. Look for periodicity by finding autocorrelations and FFT. Construct an ARIMA model to find the level of randomness.
m777182

 

[GregLocock]

Rayleigh is a bit problematical. Best bet is to rescale the x axis so that you get a symmetrical distribution, and then apply the t test etc. Alternatively go back to the mathematical derivation of the Rayleigh distribution and work out the probabilities from there. I don't know how practical that is.

You may find the chi square distribution applicable, directly to the data, but I've never used it.





Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.