نشریه بررسی های آمار رسمی ایران
سال بیست و هفتم شماره 2 (پیاپی 89، پاییز و زمستان 1395)

Nowadays, because of new methods in data collection technology, we have huge amount of data in organization. Due to important role of analyzing data and making knowledge based upon the information, using data mining techniques as an important technology for accurate and efficient data investigation, has increased. Association rule mining is as one of well-known technique in data mining science. This method is used to analyze relations between different items in datasets and discover associations’ rules in data collection as descriptive pattern.
Babor force survey is important research in occupation and Labor branch that is conducted in SCI (Statistical Centre of Iran).The key results of this survey are unemployment rate population and economic participation rate population .Recently the result show that inactive ratio population in society has increased and because of that economic participation rate has decreased. For analyzing the unknown reason, the survey dataset is investigated in order to discover relation between family member’s information and job’s status, economical activities. In this study, patterns exploration of active and inactive population is the main objective, so the Apriori algorithm as a method in association rule mining is used to determine hidden pattern in Labor force survey. The knowledge discovered in this research as descriptive rules could be used for making decision in the statistical analysis.

Many populations which are in the most focus of policy-makers are hidden or hard to reach populations that are formed by people with certain condition such as HIV infection. A careful study of these populations in order to their identification is very important, since their high risk behavior jeopardizes public health. Thus, defining a reliable and valid sampling method of them is also very vital. Applying conventional sampling methods (probability) to take samples from these populations since there is rarely a list or sampling frame of them are not practical and these methods are associated with coverage error. Non-probability sampling methods such as chain referral sampling also do not produce statistically valid estimates. One of the valid methods to take samples from these populations is respondent driven sampling method. In this sampling method, one of the most important sources of non-sampling errors is the bias which results from homophily. To check the validity of samples taken from this sampling method, in this article homophily is introduced which is a tool to assess the level of bias and the representative of samples. The main objective of this article is to examine the consequences of homophily in the production of unbiased estimators in this sampling method.

In this study, it was used logistic regression model to predictive of variables effect of education (highly educated or non-highly educated), age (young or adult), sex (male or female), location (urban or rural) and family size (low or high) on unemployment rate probability. It also was tried to analyze the changes of unemployment probability trend of these individuals during 2005 to 2013.This study was done based on the raw labor force survey results that conducted seasonality by Statistical Center of Iran (SCI) from 2005 up to now. The study results shows all variables have significant effect on unemployment probability and age, location, sex, education and family size have highest effects respectively.

Unemployment is the most controversial issue in our economy and finding solution to decline unemployment rate is the most important strategies for policy makers. In this regard attention to economic growth as a solution to decrease unemployment is taken into consideration so Okun's law is one of the achievements in this field which is explained negative relationship between unemployment and economic growth. In other words According to Okun’s law, a percentage increase in economic growth caused the unemployment rate to be reduced by 3%. In recent decades, many studies have been done on Okun's law and coefficients of Okun based on different aspects are evaluated. The result of studies on one hand indicates that the parameters, during time, vary from country to country, and region to region. In the other hand at some span of time this law has been violated. Violation of Okun's law shows a lack of proper modeling of behavior of economic variables. In this paper we use newest model named “coefficient constant and coefficient variable model” to estimate parameters of okun’s law with latest quarterly data which is released by Statistical Centre of IRAN. The results firstly indicate that unemployment has become structural unemployment. Secondly, maximum and minimum economic growth rate is 7.7 and 1.2 percent respectively.

When the purpose of sampling is to compute estimators for specific domains of population, it is possible to increase accuracy of domain estimators through stratifying population according to domains of interest and determining sample size of the domains. In this article stratified sampling design and balanced sampling design are introduced in which accuracy of domains estimators have been increased in design stage through considering domains as the strata. Also in estimating stage, utilizing modified estimators like synthetic and modified generalized regression estimator lead to more accurate estimates compare with Horvitz-Thompson estimator. Finally in a simulation study using data from 1390 establishment survey with 50 or more labor data, an application of balanced sampling approach for multi-way stratification designs for small area estimation has been explained where in synthetic estimator and modified generalized regression estimator with Horvitz-Thompson estimator for specific domains of the population are compared.

In this paper, a generalized lambda distribution is proposed followed by Lorenz, Bonferonni and Zenga curves. Also, inequality indices and Hierarchical Lorenz curve are given based on this distribution. Moreover, according to securities and exchange organization of Tehran income data, we compare some parametric Lorenz figures with their Hierarchical form based on the generalized lambda distribution.