7,237 research outputs found
Estimating risks of identification disclosure in partially synthetic data
No full textTo limit disclosures, statistical agencies and other data disseminators can release partially synthetic, public use microdata sets. These comprise the units originally surveyed, but some collected values, for example sensitive values at high risk of disclosure or values of key identifiers, are replaced with multiple draws from statistical models. Because the original records are on the file, there remain risks of identifications. In this paper, we describe how to evaluate identification disclosure risks in partially synthetic data, accounting for released information from the multiple datasets, the model used to generate synthetic values, and the approach used to select values to synthesize. We illustrate the computations using the Survey of Youths in Custody
Estimating propensity scores with missing covariate data using general location mixture models
No full textIn many observational studies, researchers estimate causal effects using propensity scores, e.g., by matching or sub-classifying on the scores. Estimation of propensity scores is complicated when some values of the covariates aremissing. We propose to use multiple imputation to create completed datasets, from which propensity scores can be estimated, with a general location mixture model. The model assumes that the control units are a latent mixture of (i)units whose covariates are drawn from the same distributions as the treated units’ covariates and (ii) units whose covariates are drawn from different distributions. This formulation reduces the influence of control units outside the treated units’ region of the covariate space on the estimation of parameters in the imputation model, which can result in more plausible imputations and better balance in the true covariate distributions. We illustrate the benefits of 1 the latent class modeling approach with simulations and with an observationalstudy of the effect of breast feeding on children’s cognitive abilities
Propensity score matching with missing covariates via iterated, sequential multiple imputation
No full textIn many observational studies, analysts estimate causal effects using propensity score matching. Estimation of propensity scores is complicated when covariate values intended for collection are in fact missing. To handle the missing data, one approach is to use multiple imputation to create completed datasets, and compute propensity scores from these datasets. However, inaccurate imputation models can result in ineffective matching, thereby limiting reductions in bias. We propose a multiple imputation approach based on chained equations in which the researcher gradually reduces the set of control units used to estimate the imputation models. This approach can reduce the influence of control records far from the treated units’ region of the covariate space on the estimation of parameters in the imputation model, which can result in more plausible imputations and better balance in the true covariate distributions. This approach can be conveniently implemented with standard multiple imputation software for missing data. Using simulations, we find that the approach can improve estimation when imputation models are mis-specified; however, it can be ineffective when imputation models are correctly specified. This suggests using the approach as part of sensitivity analysis in causal inference. We apply the approach to an observational study of the effect of breast-feeding on the child’s educational outcomes later in life
A comparison of two methods of estimating propensity scores after multiple imputation
No full textIn many observational studies, analysts estimate treatment effects using propensity scores, e.g., by matching or sub classifying on the scores. When some values of the covariates are missing, analysts can use multiple imputation to fill in the missing data, estimate propensity scores based on the m completed datasets, and use the propensity scores to estimate treatment effects. We compare two approaches to implementing this process. In the first, the analyst estimates the treatment effect using propensity score matching within each completed data set, and averages the m treatment effect estimates. In the second approach, the analyst averages the m propensity scores for each record across the completed datasets, and performs propensity score matching with these averaged scores to estimate the treatment effect. We compare properties of both methods via simulation studies using artificial and real data. The simulations suggest that the second method has greater potential to produce substantial bias reductions than the first
Pterostylis peakallana N. Reiter, B. Kosky & M. A. Clem. 2021, spec. nov.
No full textPterostylis peakallana N.Reiter, B.Kosky & M.A.Clem., spec. nov. (Fig. 7) Type: — Australia. Victoria: Wyperfeld National Park, Brocken Bucket, 130 m, 9 Nov 2019, Reiter 425 (holotype: MEL2481888 A). Pterostylis peakallana (Fig. 6, revised key to Vicflora Appendix 7) has affinities to P. exelsa but differs from that species in lacking a crest of small dense setae and a scape 18–58 cm tall (versus up to 80 cm) with 1–8 flowers (versus up to 20). Pterostylis peakallana is similar to P. terminalis but differs in labellum tip not upturned, wider lateral sepals and longer prominent setae. Pterostylis peakallana corresponds to the following illustrations: Backhouse (2019: 357; as P. species ‘north west plains’); Backhouse et al. (2016: 714; as P. ‘north west mallee’ - Yanac photos only). Deciduous terrestrial, solitary, tuberous herbs. Leaves sessile 2–10 in a rosette, imbricate, green, or withered at anthesis; lamina narrowly elliptical, margins entire, acute to acuminate, 0.5–3.5 0.5–1.0 cm. Scape 18–58 cm tall, 0.2–0.3 cm diam., generally 1–8 flowers, 3.0– 4.5 cm tall, 2.0– 2.5 cm thick. Sterile leafy bracts 4–7, ensheathing stem, lanceolate, acute, withered at flowering, 0.5–3.0 cm long. Floral bracts ovate-lanceolate, not overlapping at base of scape or at apex, clasping stem, apex acute, 2.0–3.5 0.3–1.0 cm. Pedicels erect, glabrous, slender 2–4 cm long, partially enclosed within the floral bract, appressed against the stem. Ovaries erect to slightly porrect, near but not usually against the stem, glabrous, narrowly cylindrical, 4.0–7.0 mm 1.5–2.0 mm. Flowers porrect; galea bulbous, glabrous, gibbous at base, shallowly curved throughout, narrowing to a filiform, apical point; translucent white with predominantly green to light brown stripes and markings; petals mostly translucent white with small light green or brown suffusions, margins incurved with well-developed basal flanges, usually not touching; lateral sepals usually translucent white with light brown or green outline and striping, free points usually light brown-green, basal half of lateral margins covered with prominent white trichomes. Dorsal sepal cucullate, glabrous, broadest nearest the base, 2.0– 2.5 cm long, tapering to an apical filiform, straight to decurved point, 10-14 mm long. Lateral sepals deflexed, narrow at base and conjoined in basal half, flat to shallowly concave lateral margins, sparsely ciliate, conjoined part 8–11 mm long, 7–10 mm wide; free points curved forward, filiform, 11–23 mm long, spreading. Petals asymmetrical obong-lancolate, falcate, 0.8–1.2 4.2–6.2 mm, dorsal margin prominently thickened, glaborus, anterior margin curved, glabrous, proximal flange brown prominent, apex acuminate. Labellum insectiform on a visible claw, articulate, highly sensitive to touch, nested between conjoined part of the lateral sepals when exposed and in set position; 6.2–7.9 1.7–2.9 mm, 0.3–0.4 mm deep, usually solid green or brown, laminar oblanceolate with basal lobe: 1.6–2.5 0.7–1.3 mm, 1.4–2.0 mm wide; distal margins slightly uneven with a series of spreading white setae, largest nearest the base, smaller towards apex, 1.1–2.5 mm long, with twin pairs of prominent long setae, 2.6–4.6 mm long below the basal lobe, occasionally a set of hairs, 1.0– 1.5 mm, extending from the basal mound, not always present. Column porrect, incurved, extending most of the length of the bulbous part of the galea, partly visible through translucent parts of the galea; column wings rectangular, distal margins covered in a series of fine, dense setae. Anther obtuse, green, c. 1 mm long. Pollinia linear, yellow. Flowering: — Peak flowering November. Habitat: — Different semi-arid habitats, the first a floristically depauperate, low-lying, seasonally inundated open Eucalyptus woodland on cracking clays dominated by Eucalyptus calycogona subsp. trachybasis and Eucalyptus phenax subsp. phenax, and the second site a diverse open eucalypt woodland dominated by Eucalyptus largiflorens, Bursaria spinosa Cavanilles (1797: 30), Chrysocephalum semipapposum Steetz in Lehman (1845: 474), Rhodanthe corymbiflora (Schlechtendal 1848: 448) Wilson (1992: 391) with a diverse understory community of grasses and geophytes. Distribution: — Currently known from two widely disjunct populations (Fig. 1) within Wyperfeld National Park and a reserve near Boort, with unconfirmed reports from one further location within Wyperfeld National Park. Conservation status: — Endangered, P. peakallana meets the IUCN criteria (IUCN, 2012) EN D for endangered with less than 250 individuals known. However, considering the large area of similar habitat within Wyperfeld, Murray Sunset and surrounding reserved areas, adequate surveys for this species in mid-November in years of high rainfall are required to accurately determine its status. Etymology:— Named in honour of Rod Peakall in recognition of his role in elucidating the pollination mechanisms of many Australian orchids and his contributions to understanding evolution of pollination by sexual deception. Additional specimens examined:— Australia. Victoria: Wyperfeld National Park, Brocken Bucket, 130 m, 9 Nov 2019, Reiter 426 (MEL2481889 A); Wyperfeld National Park, Brocken Bucket, 130 m, 9 Nov 2019, Reiter 423 (MEL2481886 A); Wyperfeld National Park, Brocken Bucket, 130 m, 9 Nov 2019, Reiter 424 (MEL2481887 A); Marmal NCR, 9 Nov 2019, Reiter 420 (MEL2481883 A); Marmal NCR, 9 Nov 2019, Reiter 422 (MEL2481885 A); Marmal NCR, 104 m, 9 Nov 2019, Reiter 421 (MEL2481884 A); Marmal NCR, 104 m, 3 Nov 2016, Radford s.n. (MEL2411314); Marmal NCR, 104 m, 9 Nov 2019, Reiter 433 (MEL2481896 A).Published as part of Reiter, Noushka, Kosky, William & Clements, Mark, 2021, Two new species of Pterostylis (Orchidaceae; Orchidoideae) from the Sunset Country, Victoria, Australia, pp. 153-165 in Phytotaxa 500 (3) on pages 162-163, DOI: 10.11646/phytotaxa.500.3.1, http://zenodo.org/record/542454
Semi-supervised classification of injection moulding processes
No full textAuthor DI Oliver Reiter, BScMasterarbeit Universität Linz 2023Arbeit nach Ablauf der Sperre auf den öffentlichen PCs in den Bibliotheken der JKU+Medizin abrufba
- …
