THE IRAQI MAGAZINE FOR ADMINISTRATIVE SCIENCES,
2017, Volume 13, Issue 51, Pages 278-298
AbstractThis paper discusses the reliance of numerical analysis on the concept of the standard deviation (SD) which is widely used in Different statistical research, it’s very important in the statistical analytical. Most importantly from that, however many traditional statistics depend on it, such as (F test, analysis of variance, the effect sizes, ets). Considered the mean deviation (MAD) of the significant measures of Dispersion and reasonable alternatives competition for standard deviation (SD) and it has many uses. In this research, conducted compared historical between (SD) and (MAD) as it mention his research by the researcher Eddington  confirm parameter efficiency (MAD) and researcher Fisher  supported the efficiency parameter (SD), and supporters and opponents of each of them. Moreover, here we would like to mention that the statistical Verdict for (SD), who confirmed his support full superiority by the researcher Fisher hadn't always have the best. But we argued here, that the absolute mean deviation (AMD), has many advantages over the standard deviation, he is more efficient as an estimate of a population parameter in the real-life situation where the data contain tiny errors, or do not form a completely perfect normal distribution. Finally the research adopted technique simulation and Real data to compare (SD), and (MAD) using a normal distribution of each of them and an upnormal distribution of each of them. .
- Article View: 30
- PDF Download: 143