Search Results

This chapter describes research on identification, whose aim is credible use of evidence to inform patient care. Methodological research should aim to determine the information relevant to patient care that studies provide, when evidence is combined with credible assumptions. Statistical imprecision and identification problems limit the informativeness of studies. It also stems from small sample size. Identification problems are the difficulties that persist when sample size grows without bound. Statistical imprecision may be vexing, but identification is the more fundamental challenge. As such, this chapter begins by describing common identification problems that arise when studying trial or observational data. The chapter then discusses econometric research that analyzes these problems.

This chapter takes the scattered price data that have survived in order to make an index of Roman inflation. The index cannot describe the details of inflation, but it allows comparison between periods of time. The chapter suggests that the persistent inflation of the late Roman Empire came from political instability at the top of the hierarchy. It also raises an important problem in going from correlation to causation—the identification problem. As a result, economists propose an identification strategy to suggest how Roman inflation got its start in the late second century. The chapter also shows how, despite a great scarcity of price data for the Roman Empire, ancient historians appear to have conventional opinions about Roman price movements.

The topic of this chapter is testing linearity of a model when the alternative is a parametric nonlinear model nesting a linear model. The case leading to standard asymptotic inference is discussed first. This is followed by a characterization of the identification problem present in many of these testing situations. Different ways of solving the identification problem are considered and their merits and disadvantages discussed. The chapter also covers tests designed for situations in which no well‐specified alternative to the linear model in question is available. Finally, there is a discussion of the concept asymptotic relative efficiency that is helpful in comparing power properties of various tests in different testing situastions and brief remarks on which test or tests to use in practice and when.

This chapter discusses empirical methodology, focusing on attenuation bias induced by measurement error or data fluctuation and solutions to the biased and inconsistent empirical estimates of the autoregression model. It shows that the identification problem arises in large part due to the non-observability of genotypes. Even if genotypes can be observed, an identification problem in observational data might still arise because genotypes and outcomes, such as wages, schooling, mothering, and infant distress, are jointly determined. One solution is to exploit panel data, which in principle exist for generations of the same family (greatgrandchildren, grandchildren, children, parents, grandparents, greatgrandparents). A number of methodologies may be used to solve the identification problem, including the instrumental-variables estimator. The chapter also presents a methodological critique of behavioral genetics and genome-wide association studies and describes the equal-environments assumption. It argues that behavioral genetics cannot identify the role of heredity in the correlation between siblings.

This chapter examines relevant research about hidden organizations as it has emerged in various disciplines. It addresses work on secret societies, criminal organizations, the informal economy, terrorist organizations, secret government agencies, and other hidden collectives. Several conclusions are drawn from this research, which are helpful in talking about the strategic communication of organizational and organizational member identity in these hidden organizations.

This chapter discusses what happened to the Polish Jewish captives once they had been ransomed and released. Most sought to return home at the first opportunity. Without a patron, however, this was not easy. The captives had been brutally snatched from their previous lives and so, once freed, had little or no resources on which to rely. The religious obligation of the local Jewish society toward them ended with their ransom; once they had been freed, they were largely on their own and had to make their own way home—an extremely difficult, often impossible, proposition. There is no way to tell just what proportion of the ransomed captives managed to return home, though the desire to do so seems to have been fairly widespread. Still, there were those who simply could not manage it. The financial difficulties, the physical danger of long-distance travel, and the continuing threat of pirates in the Mediterranean must have deterred many, especially women, who often opted to stay and start new lives. The chapter then considers the refugee information network, the problems of identification, and the cultural contacts between Ashkenazi refugees and the Sephardi society.

We revisit the twofold identification problem discussed by Cho, Ishida, and White (2011), which arises when testing for neglected nonlinearity by artificial neural networks. We do not use the so-called “no-zero” condition and employ a sixth-order expansion to obtain the asymptotic null distribution of the quasi-likelihood ratio (QLR) test. In particular, we avoid restricting the number of explanatory variables in the activation function by using the distance and direction method discussed in Cho and White (2012). We find that the QLR test statistic can still be used to handle the twofold identification problem appropriately under the set of mild regularity conditions provided here, so that the asymptotic null distribution can be obtained in a manner similar to that in Cho, Ishida, and White ( 2011). This also implies that the weighted bootstrap in Hansen (1996) can be successfully exploited when testing the linearity hypothesis using the QLR test.

This chapter discusses the settlement evolution in the territory of Sagalassos (south-west Turkey) from the start of the Byzantine period until the thirteenth century when Sagalassos was ultimately abandoned and habitation moved to new locations in and around the modern village of Ağlasun. Problems regarding the archaeological recognition characterize the Byzantine material culture of the region. Recent excavations at Sagalassos together with focused ceramic studies and ongoing intensive surveys are changing this and providing insights into a history of habitation that is not uniform within the territory and that is sometimes at odds with processes occurring in Anatolia at large.