Non-ignorable missing covariate data in survival analysis: a case-study of an International Breast Cancer Study Group trial

Summary.  Non-ignorable missing data, a serious problem in both clinical trials and observational studies, can lead to biased inferences. Quality-of-life measures have become increasingly popular in clinical trials. However, these measures are often incompletely observed, and investigators may suspect that missing quality-of-life data are likely to be non-ignorable. Although several recent references have addressed missing covariates in survival analysis, they all required the assumption that missingness is at random or that all covariates are discrete. We present a method for estimating the parameters in the Cox proportional hazards model when missing covariates may be non-ignorable and continuous or discrete. Our method is useful in reducing the bias and improving efficiency in the presence of missing data. The methodology clearly specifies assumptions about the missing data mechanism and, through sensitivity analysis, helps investigators to understand the potential effect of missing data on study results.     

Wealth Inequality in Black and White: Cultural and Structural Sources of the Racial Wealth Gap

Using data from the 2013 Survey of Consumer Finances, this research examines competing and complementary cultural and structural explanations of the sources of racial differences in wealth. We use OLS regression and quantile regression to identify the major individual-level sources of wealth differences between African Americans and whites. Whites have more favorable wealth characteristics than do African Americans on all of the variables in the analysis: gender of household head, bankruptcies, spending patterns, stock ownership, business ownership, home ownership, inheritance, educational attainment, income, occupation, age, and number of children. Cultural factors, having a female-headed family, spending patterns, and inheritance account for little of the racial wealth gap. Racial differences in income, stock ownership, and business ownership account for much of the explained racial wealth gap. Moreover, compared with whites, African Americans receive significantly lower wealth returns to education, age, income, stock ownership, and business ownership. We discuss the implications of our findings.     

The robust economic statistical design of the Hotelling's T2 chart

ABSTRACT

Economic statistical designs aim at minimizing the cost of process monitoring when a specific scenario or a set of estimated process and cost parameters is given. But, in practice the process may be affected by more than one scenario which may lead to severe cost penalties if the wrong design is used. Here, we investigate the robust economic statistical design (RESD) of the T² chart in an attempt to reduce these cost penalties when there are multiple scenarios. Our method is to employ the genetic algorithm (GA) optimization method to minimize the total expected monitoring cost across all distinct scenarios. We illustrate the effectiveness of the method using two numerical examples. Simulation studies indicate that robust economic statistical designs should be encouraged in practice.     

Longitudinal Consistency in Self-Reported Age of First Vaginal Intercourse among Young Adults

We examined consistency in self-reports of age at first vaginal sex among 9,399 male and female respondents who participated in Waves III and IV (separated by approximately seven years) of the National Longitudinal Study of Adolescent Health (Add Health). Respondents were coded as consistent if they reported an age at first vaginal intercourse at Wave IV that was within one year of the age they reported at Wave III. Sociodemographic, behavioral, and cognitive predictors of consistency were examined using bivariate and multivariate logistic regression. Overall, 85.43% of respondents were able to provide consistent reports. Among both males and females, consistency was associated with age, years since first vaginal intercourse, race/ethnicity, and lifetime number of other-sex partners in final multivariate models. Respondents who were older and had more recently had their first sexual experience were more likely to be consistent. For females only, those who reported a history of nonparental, physically forced sex were less likely to be consistent. Most young adults consistently report age at first vaginal intercourse, supporting the credibility of retrospective self-reports about salient sexual events such as timing of first vaginal intercourse.     

The Right to Take Risks: Alcohol and Older People

This paper examines policy and practice regarding the purchase of alcohol for older clients of home carers in three local authorities in the Greater London area. Data were gathered from interviews with home carers and their managers, focus group discussions, and a postal survey. None of the local authorities had a written policy or written guidance on alcohol purchase or on appropriate responses to the identification of problematic drinking in older clients. The findings illustrate the problem of balancing "rights" and "risks" within a philosophy of community care which emphasizes client choice and autonomy, and show how policy and practice are "tailored" by local contexts. It is suggested that both home carers and other domiciliary workers, as well as their older clients, could benefit from a clear code of practice regarding responses to alcohol consumption and problematic drinking by older people.     

The relationship between criminality and home background

There are two extensive literatures which deal with this subject. One is that on criminality, defining the distinctions between morality and cultural definitions. The other is on how parents ought to influence their offspring. And there is also a smaller literative on the causal or statistical relationship between the two. This research tradition attempts to isolate particular hypotheses—like lone parents or genetics. What is missing is both a study of why certain influences are significant, and a connection between the actual cases of criminality and criminals' own experiences of their parents. For this purpose lengthy semistructured interviews with young offenders were carried out, exploring their memories and perceptions of home life and their attitudes to their parents and their subsequent experiences. This paper discusses the factors that the interviewees raised consistently, openly and articulately; violence in the home, neglect, laissez-faire: and turning to peer groups for alternative support. At the heart of the evidence lies a lack of dialogue and the perception of a lack of active interest in them by parents. The consequences are discussed, including the inability to develop normal social relationships, leading to frustration and the repetition of violence in the home in different circumstances.     

On the use of curve fitting to model the error of the Cornish-Fisher expansion of the pearson type-VI distribution

The Cornish-Fisher expansion of the Pearson type VI distribution is known to be reasonably accurate when both degrees of freedom are relatively large (say greater than or equal to 5). However, when either or both degrees of freedom are less than 5, the accuracy of the computed percentage point begins to suffer; in some cases severely. To correct for this, the error surface in the degrees of freedom plane is modeled by least squares curve fitting for selected levels of tail probability (.025, .05, and .10) which can be used to adjust the percentage point obtained from the usual Cornish-Fisher expansion. This adjustment procedure produces a computing algorithm that computes percentage points of the Pearson type VI distribution at the above probability levels, accurate to at least + 1 in 3 digits in approximately 11 milliseconds per subroutine call on an IBM 370/145. This adjusted routine is valid for both degrees of freedom greater than or equal to 1.