This paper deals with the ANALYSIS of log-linear models for multi-way contingency TABLES utilizing the empirical Bayes procedure. We consider such a table as a multi-nomial distribution involving a vector of parameters.A log-linear representation is introduced. Assuming a Dirichlet prior for the parameters of the log-linear model, we employ the Bayesian paradigm to infer the structure of the table. Assuming the exact prior distribution is not known, the empirical Bayes procedure is utilized to estimate the prior. Based on the estimated posterior, the Bayes estimator and its distributional properties are studied. Asymptotic normality of the estimators is established. The procedure is illustrated by an example.

This experiment was conducted to determine the effect of nutrient ANALYSIS TABLES of ingredient on Ross strain broiler performance. This experiment used CRD design. Chicks were grouped in to three treatments with six replications and each with 15 chicks. The experimental treatment consisted: 1) Diet formulation according to ANALYSIS table of NRC. 2) Diet formulation according to ANALYSIS table of FEEDSTUFF and 3) Diet formulation according to ANALYSIS table of DEGUSSA. Feed intake (FI), Average daily gain (ADG), Feed conversion ratio (FCR) and Production index (PI) were measured. The cost of feed per1kg of weight and carcass characteristics were also compared. According to the result the body weight among different treatments was significant (p<0.05). The highest final weight was obtained with NRC ANALYSIS table. Feed intake among treatments was not significantly difference (p>0.05). There was significant difference in feed conversion among treatments and NRC ANALYSIS table was resulted the lowest FCR. Production index among treatments was significantly difference (p<0.05). The cost of feed per 1kg weight among treatment was not significant (p>0.05).Finally, with respect of all parameters, the best treatment was NRC ANALYSIS table.

The existing studies on the status of age pattern of marriage of Iranian women are mainly based on various hypotheses such as marriage squeeze and changes in the marriage market. However, by increasing the share of educated women, education can affect the age pattern and delay in women's marriage independently. Emphasizing on the effect of education on the age pattern of marriage, this article attempts to evaluate average changes in the years of celibacy of women using cohort and cross-sectional ANALYSIS. For this purpose, cross-sectional data were generated for the 2016 and cohort data for three hypothetical cohorts born in the periods 1981-86, 1986-91 and 1991-96 by educational levels and urban and rural areas. Then, using the Cole's proposed method, the cohort experience of women's marriage was completed for all three cohorts and the cohort and cross-sectional marriage ratios by education and urban-rural areas were entered into the nuptiality life table. The findings indicate that the highest average years of celibacy are related to university women, so that the cohort born in 1986-91 will experience the longest delay and the cohort born in 1991-96 will experience the least delay in marriage. The comparison of urban and rural data indicates that with an increase in level of education, rural women will experience more delay in marriage than urban women.

A hierarchical Bayes model for ANALYSIS of contingency TABLES is presented and, using it, inference about correlation parameter, logarithm of the odds ratio, is studied. For drawing random sample from the distribution of logarithm of the odds ratio, computational method of Gibbs sampling is used. For testing independence the use of the hierarchical Bayes in calculating the Bayes factor is introduced and in an applied example is employed.

Objective: The nutrition of children in kindergartens has a great significance for proper growth and development of children. In order to save time and money, the control of macronutrients content is performed by calculations using food composition TABLES instead of performing a chemical ANALYSIS.Methods: We examined the macronutrients content of 240 whole day meals using food composition TABLES and performed chemical ANALYSIS of meals to determine adequacy and validity of food composition TABLES in calculation of macronutrient contents in kindergarten meals.Findings: We established no correlation (P>0.05) between the value of proteins, fats and carbohydrates. Significant difference was established between the average content of proteins (t=2.57; P<0.05), and carbohydrates (t=3.20; P<0.01), but not with the content of fats in the meals (t=1.26; P>0.05) (food composition TABLES vs chemical ANALYSIS).Conclusion: Until we establish new food composition TABLES, chemical ANALYSIS remains the only valid method for assessment of macronutrients content and energy value of a meal in kindergarten.