Home › Forums › Old Forums › General › about the random generator

This topic contains 3 replies, has 4 voices, and was last updated by BTDT 11 years, 3 months ago.

Viewing 4 posts - 1 through 4 (of 4 total)

- AuthorPosts
- January 2, 2007 at 9:08 pm #119326
along with the minitab software, i generated 100 random data from a normal distribution with default(mean=0,StDev=1) and posted along C1. then i went to the menus. i clicked stat followed by basic stat and then descriptive statistics….i noticed that the result of the mean = 0.112 and StDev = 1.030 along with C1, which is not the same with the default(mean=0,StDev=1), and why is this so? when in fact the data generated are derived from mean=0 and StDev=1. is there something you can say about the random generator? or i need a clear answer plz…

January 2, 2007 at 9:18 pm #119329I’m not a Minitab representative but let me explain my understanding of their random function after using it for many years. Each data point is randomly selected and not dependent on the other points previously selected. This allows the mean and standard deviation to be close but not exactly equal to the criteria used to generate your rows of data.

If you use more points, e.g. 1000 than 100, you will find the generated sample’s mean and standard deviation are closer to your criteria. It is easier to have a few more points randomly from the edges of the distribution affect a smaller number of points than a larger data set.

Hope this helps.January 2, 2007 at 9:23 pm #119331You took a sample of 100. The true population size is infinite. 100 does not equal infinity. The results of the t and the F test are

t = (.112 – 0)/(1.03/sqrt(100)) = 1.08

F = (1.03*1.03)/1*1 = 1.061

for 99 df a t of 1.08 at 95% is not signficant. For an F test with df = 100 and infinity an F of 1.061 is not significant at 95% therefore your sample mean is not significantly different from 0 and your sample standard deviation is not significantly different from 1.

Your random generator is working fine.January 2, 2007 at 10:11 pm #119337Euge:The large sample properties of the pseudo-random numbers from Minitab will satisfy all your requirements. If you wish to repeat the selection on a number of different laptop computers at different time with the same result, specify a particular number in the ‘seed’ box.The parameters from a sample (average, range, standard deviation, etc.) will show a scatter in the same manner as the uniform pseudo-random numbers from Excel using the “=rand()” function.If you throw a pair of dice a large number of times, the long term average will approach seven, but you would not expect the average of three throws to be exactly seven each time.There is nothing wrong with the pseud-random number generator in Minitab. You are using it correctly.Cheers, BTDT

- AuthorPosts

Viewing 4 posts - 1 through 4 (of 4 total)

The forum ‘General’ is closed to new topics and replies.

© Copyright iSixSigma 2000-2018. User Agreement. Any reproduction or other use of content without the express written consent of iSixSigma is prohibited. More »