مقارنة بين بعض طرائق تقدير حجم العينة لتقدير معلمات أنموذج التصنيف في حالة وجود بيانات عالية الابعاد باستعمال المحاكاة

Abstract EN

In this research, several methods were used to determine the optimum sample size to estimate the parameters of high-dimensional data (HDD).

Where the number of explanatory variables is greater than the sample size (P> n).

These methods are the Bonferroni inequality method a special case of normal approximation and the Bernstein's inequality method.

The non-linear logistic regression model is estimated in the sample size of each method in the case of high-dimensional data using the artificial intelligence method, which is the artificial neural network method (ANN), as it gives a high-precision estimate commensurate with the data type and type of medical study.

The probabilistic values obtained from the artificial neural network are used in the calculation of the net reclassification index.

A program was written for this purpose using the statistical programming language (R) where the mean maximum olute error criterion (MME) of the net reclassification network index (NRI) was used to compare the methods of specifying the sample size and the presence of a number of different default parameters under the given margin of error value (ε).

To verify the performance of the methods using the comparison criteria above where the most important conclusions were that the Bernstein's inequality method is the better in determining the optimal sample size according to the number of default parameters and the error margin value.