Search

We consider sensitivity of a generic stochastic optimization problem to model uncertainty. We take a non-parametric approach and capture model uncertainty using Wasserstein balls around the postulated model. We provide explicit formulae for the first-order correction to both the value function and the optimizer and further extend our results to optimization under linear constraints. We present applications to statistics, machine learning, mathematical finance and uncertainty quantification. In particular, we provide an explicit first-order approximation for square-root LASSO regression coefficients and deduce coefficient shrinkage compared to the ordinary least-squares regression. We consider robustness of call option pricing and deduce a new Black-Scholes sensitivity, a non-parametric version of the so-called Vega. We also compute sensitivities of optimized certainty equivalents in finance and propose measures to quantify robustness of neural networks to adversarial examples., (© 2021 The Authors.)

Objective: To evaluate the adherence to asthma evidence-based guidelines before and after a quality improvement process., Design: A controlled trial was conducted at two regional hospitals (intervention and control hospitals). We performed a retrospective pre-intervention audit, followed by a post-intervention audit 1 year after the implementation of evidence-based guidelines., Setting: Emergency departments of two neighbouring hospitals serving regional and rural North West Tasmania., Participants: All children (<18 years) with acute presentation of doctor-diagnosed asthma., Interventions: Implementation of evidence-based guidelines using the National Asthma Council of Australia and the Global Initiative of Asthma guidelines, at the intervention hospital and care as usual at the control hospital., Main Outcome Measure(s): The main outcome measures were the compliance to evidence-based guidelines, pre- and postintervention at the intervention hospital, compared to the control hospital. The specific outcomes measure included the clinical presentation, management, referral to asthma and allergy clinic, and hospitilisation., Results: Significantly improved adherence to evidence-based guidelines were noted post-intervention at the intervention hospital, that is severity recorded (21.4%-45.7%, P < 0.001), triggers identified (13.5%-45.3%, P < 0.001), spirometry usage (3.8%-25.8%, P = 0.03) and written action plans (29.7%-58.3%, P < 0.001). There was however no effect on hospitilisation (23.3%-29.8%, P = 0.48). At the control hospital, however, no significant improved adherence to evidence-based guidelines were noted., Conclusions: Evidence-based implementation led to improved adherence to evidence-based guidelines across an expanded list of domains in a regional setting., (© 2019 National Rural Health Alliance Ltd.)