![]() ![]() ![]() The Kolmogorov-Smirnov test is available in some statistical software. The test statistic is compared against the critical values from a normal distribution in order to determine the p-value. For each test discussed below, the formal hypothesis test is written as: So this section provides a discussion of some common testing procedures (of which there are many) for normality. However, more rigorous and formal quantification of normality may be requested. Typically, assessment of the appropriate residual plots is sufficient to diagnose deviations from normality. ![]() It means that it is reasonable to assume that the errors have a normal distribution. A large p-value and hence failure to reject this null hypothesis is a good result. To complement the graphical methods just considered for assessing residual normality, we can perform a hypothesis test in which the null hypothesis is that the errors have a normal distribution. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |