On the estimation of a sensitive quantitative mean using blank cards

In this article, we introduce for the first time, the blank card methods for estimation of finite population mean of a sensitive variable. Two generic randomization devices are suggested, and for each device we identify the choices of special models. We introduce additive, multiplicative, and combination of both additive and multiplicative scrambling models that require use of a non sensitive variable. We derive the basic statistical properties of each model. It is interesting to note that various existing estimators can be viewed as the special cases of those presented here. The statistical efficiency of proposed techniques is compared with Greenberg et al. (1971 Greenberg, B.G., Kuebler Jr, R.R., Abernathy, J.R., Horvitz, D.G. (1971). Application of the randomized response technique in obtaining quantitative data. J. Am. Stat. Assoc. 66(334):243–250.[Taylor &; Francis Online], [Web of Science ®] , [Google Scholar]) and modified Perri (2008 Perri, P.F. (2008). Modified randomized devices for Simmons’ model. Model Assisted Stat. Appl. 3(3):233–239. [Google Scholar]) model. The proposed devices can easily be adjusted to achieve the required efficiency level by making suitable choices of different design parameters.     

The Robustness of High Danish National Happiness: A Temporal Cross-Country Analysis of Population Subgroups

Denmark’s top position in various rankings of country happiness is well-documented. This study goes beyond the national average comparisons and investigates whether Denmark’s top position is also found when we disaggregate data in line with social categories often used within the social sciences. The central measure is the empirical probability that a given population subgroup in Denmark has significantly higher happiness compared to another country’s similar subgroup in a given year. All five rounds of the European Social Survey are used but only the sixteen countries that were surveyed in each of the five rounds are included in this study. The results show that Denmark’s position at the top of the happiness scale is also robust when we look at population subgroups, but not in the sense that Denmark dominates all countries for all years. Instead, a modified version of robustness is necessary; Denmark very often has significantly higher happiness levels than in other countries, only sometimes has the same happiness levels as in other countries, very rarely is it dominated by other countries, and finally it is never dominated by other countries in all 5 years for a given subpopulation characteristic. This conclusion is quite insensitive to the applied SWB measure and the applied significance level.

     

Improved randomized response in additive scrambling models

Randomized response models deal with stigmatizing variables appearing in health surveys. Additive and subtractive scrambling in split sample and double response yield unbiased mean and sensitivity estimators of high precision. The split sample method is protective of privacy. The double response method is as protective only conditionally. To achieve the maximum efficiency, the scrambling variables must be similar to each other and the probability of obtaining a true response must be as large as possible. The randomized response procedures yield more efficient estimates of the average total number of classes missed by university students.     

The sensitivity of income polarization

This study looks at polarization and its components’ sensitivity to assumptions about equivalence scales, income definition, ethical income distribution parameters, and the income accounting period. A representative sample of Danish individual incomes from 1984 to 2002 is utilised. Results show that polarization has increased over time, regardless of the applied measure, when the last part of the period is compared to the first part of the period; primary causes being increased inequality (alienation) and faster income growth among high incomes relative to those in the middle of the distribution. Increasing the accounting period confirms the reduction in inequality found for shorter periods, but polarization is virtually unchanged, because income group identification increases. Applying different equivalence scales does not change polarization ranking for different years, but identification ranks are affected. The welfare state considerably reduces income polarization and inequality, but at the expense of some more identification.      

Securitized citizens: Islamophobia,racism and the 7/7 London bombings

The London bombings of 7 July 2005 were a major event shaping the relationship between Muslims and non‐Muslims in Britain. In this paper we introduce the idea of ‘securitized citizens’ to analyse the changing relationship between British Muslims and wider British society in response to this and similar events. Through an analysis of qualitative interviews with Muslims and non‐Muslims of a variety of ethnic backgrounds in the areas where the London bombers lived in West Yorkshire we examine the popular perceptions of non‐Muslims and Muslims’ experiences. We show how processes of securitization and racialization have interacted with Islamophobic discourses and identifications, as well as the experiences of Muslims in West Yorkshire after the attacks.     

Demographic crisis: The impact of the Bangladesh civil war (1971) on births and deaths in a rural area of Bangladesh

Summary In Matlab Bazaar Thana the Cholera Research Laboratory has registered the births, deaths and migrations in a population of approximately 125,000 since 1966. Although this rural area was not the scene of any significant armed encounters, striking changes in birth and death rates were registered during and after the conflict. Birth rates did not change during the relatively brief period of the civil war, but a small decline was registered for one year after the war. Fertility rates which had been declining slightly and irregularly in the pre-war baseline period may have increased slightly during the war and fell substantially in all age groups in the year following the war. The crude death rate, which rose by 37 per cent during the war, was a very sensitive reflection of the administrative and economic problems. Overall infant mortality rose by only 15 per cent over pre-war levels because all of the increase was observed in the post-neo-natal component, which traditionally accounts for less than one-third of the total infant mortality in Bangladesh. Children and older adults accounted for the majority of excess deaths which were largely attributed to acute diarrhoeas and other gastro-intestinal causes. The death rate at ages 1-4 rose by 43 per cent and at ages 5-9 soared to 208 per cent above pre-war baseline rates. All increases in age-specific mortality rates fell to baseline levels during the year following the war, except the 5-9-year age group, in which rates continued to be high largely because of deaths due to dysentery.     

Choosing summary statistics by least angle regression for approximate Bayesian computation

Bayesian statistical inference relies on the posterior distribution. Depending on the model, the posterior can be more or less difficult to derive. In recent years, there has been a lot of interest in complex settings where the likelihood is analytically intractable. In such situations, approximate Bayesian computation (ABC) provides an attractive way of carrying out Bayesian inference. For obtaining reliable posterior estimates however, it is important to keep the approximation errors small in ABC. The choice of an appropriate set of summary statistics plays a crucial role in this effort. Here, we report the development of a new algorithm that is based on least angle regression for choosing summary statistics. In two population genetic examples, the performance of the new algorithm is better than a previously proposed approach that uses partial least squares.     

Factor screening in nonregular two-level designs based on projection-based variable selection

In this paper, we focus on the problem of factor screening in nonregular two-level designs through gradually reducing the number of possible sets of active factors. We are particularly concerned with situations when three or four factors are active. Our proposed method works through examining fits of projection models, where variable selection techniques are used to reduce the number of terms. To examine the reliability of the methods in combination with such techniques, a panel of models consisting of three or four active factors with data generated from the 12-run and the 20-run Plackett–Burman (PB) design is used. The dependence of the procedure on the amount of noise, the number of active factors and the number of experimental factors is also investigated. For designs with few runs such as the 12-run PB design, variable selection should be done with care and default procedures in computer software may not be reliable to which we suggest improvements. A real example is included to show how we propose factor screening can be done in practice.     

Testing for autocorrelation in non-stationary dynamic systems of equations

Using Monte Carlo methods, the properties of systemwise generalizations of the Breusch-Godfrey test for autocorrelated errors are studied in integrated cointegrated systems of equations. Our analysis, regarding the size of the test, reveals that the corrected LR tests have been shown to perform satisfactorily even in cases when the exogenous variables follow a unit root process, whilst the commonly used TR2 test behaves badly even in single equations. All tests perform badly, however, when the number of equations increases and the exogenous variables are highly autocorrelated.