Learning Financial Networks with High-frequency Trade Data.

Financial networks are typically estimated by applying standard time series analyses to price-based economic variables collected at low-frequency (e.g., daily or monthly stock returns or realized volatility). These networks are used for risk monitoring and for studying information flows in financial markets. High-frequency intraday trade data sets may provide additional insights into network linkages by leveraging high-resolution information. However, such data sets pose significant modeling challenges due to their asynchronous nature, complex dynamics, and nonstationarity. To tackle these challenges, we estimate financial networks using random forests, a state-of-the-art machine learning algorithm which offers excellent prediction accuracy without expensive hyperparameter optimization. The edges in our network are determined by using microstructure measures of one firm to forecast the sign of the change in a market measure such as the realized volatility of another firm. We first investigate the evolution of network connectivity in the period leading up to the U.S. financial crisis of 2007-09. We find that the networks have the highest density in 2007, with high degree connectivity associated with Lehman Brothers in 2006. A second analysis into the nature of linkages among firms suggests that larger firms tend to offer better predictive power than smaller firms, a finding qualitatively consistent with prior works in the market microstructure literature.

Respecting the Patient's Choice: A Case of Possible Drug-Induced Parkinsonism.

This report describes a case of likely drug-induced Parkinsonism (DIP) identified by the pharmacist. A 54-year-old female patient was referred by a physician to the pharmacist in a rural, integrated care team for a comprehensive medication review (CMR) to address the patient's concerns of possible Parkinson's disease (PD). While PD may occur over the progression of age, medications that affect dopamine transport can also cause DIP, a secondary form of Parkinson's disease. Although PD and DIP may be clinically indistinguishable, differentiation may be possible by reviewing a patient's medication history for any potential causative drugs correlating to the timeline of the onset of symptoms. In this case, the pharmacist reviewed the medication profile and identified medications that could be responsible for causing DIP, specifically bupropion. The pharmacist suggested discontinuing bupropion and identifying another option for treating depression. The patient appreciated the suggestion and education, but ultimately preferred continuing her bupropion therapy instead of discontinuing therapy or changing to an alternative agent. At a follow-up meeting with the pharmacist, not only was the patient still experiencing tremors despite taking carbidopa/levodopa, but additional medications known to be potential inducers of tremors were added to her regimen. Although the pharmacist repeatedly discussed DIP with the patient and believed stopping bupropion would determine whether her Parkinsonism was PD or DIP, ultimately the patient continued taking bupropion because of concerns related to depression severity and the impact on her well-being. The patient's wishes were respected.

A case of pharmacist-led care team interventions to maximize rural patient quality of life.

A 64-year-old rural home bound patient in Washington State was identified to be at increased risk for negative health care outcomes related to chronic end kidney disease and poorly controlled diabetes. The patient lacked understanding of the use of monitoring equipment as well as diabetes education to improve quality of health; he also did not have access to medical supply equipment. A pharmacist-led care team comprised of a pharmacist, a community health worker, and a home health nurse implemented comprehensive medication review techniques as well as direct patient care education to engage the patient in managing his health. Involvement with this home visit care team combined with the patient's recently developed interest in managing his health re-engaged the patient. He began attending more frequent visits with his providers and increased his interest in meeting with a diabetes educator at the local clinic. Resulting interactions with the patient's providers, pharmacy, and community resources increased patient's compliance, access to specialists of care, and in-home safety measures. Factors contributing to poorer overall health and higher rates of death among rural patients include increased travel time to health care facilities and providers, higher rates of unhealthy lifestyle choices such as cigarette smoking and obesity, higher rates of poverty and less access to healthcare in general. This scenario emphasizes the important role an interprofessional team plays in the care of isolated, rural health patients in managing chronic disease states for stability as well as quality of life.

Multidimensional diffusion processes in dynamic online networks.

We develop a dynamic matched sample estimation algorithm to distinguish peer influence and homophily effects on item adoption decisions in dynamic networks, with numerous items diffusing simultaneously. We infer preferences using a machine learning algorithm applied to previous adoption decisions, and we match agents using those inferred preferences. We show that ignoring previous adoption decisions leads to significantly overestimating the role of peer influence in the diffusion of information, mistakenly confounding influence-based contagion with diffusion driven by common preferences. Our matching-on-preferences algorithm with machine learning reduces the relative effect of peer influence on item adoption decisions in this network significantly more than matching on earlier adoption decisions, as well other observable characteristics. We also show significant and intuitive heterogeneity in the relative effect of peer influence.

Cholestasis secondary to panhypopituitarism in an infant.

Cholestasis occurring in infancy should be evaluated completely to exclude hepatic as well as endocrine or metabolic causes. A rapid diagnosis should be made to ensure that hepatic and neurologic complications are prevented. We describe a rare case of infant cholestasis due to panhypopituitarism resulting in hypoglycemia and liver biopsy findings consistent with bile duct paucity. Existing bile ducts were noted to be small, and electron microscopy demonstrated diminutive and atrophied biliary cells with a diminished bile duct lumen size. Hypoglycemia and cholestasis resolved with treatment of the underlying panhypopituitarism. Panhypopituitarism should be considered in any infant who presents with cholestasis, hypoglycemia, and other manifestations of pituitary malfunction. Growth hormone deficiency may affect bile duct formation as demonstrated in this patient.

Previously unsuspected infantile hypertrophic pyloric stenosis diagnosed by endoscopy.

Infantile hypertrophic pyloric stenosis is a common cause of upper gastrointestinal obstruction in infants and presents as projectile, nonbilious emesis in the first few weeks of life. In general, patients with infantile hypertrophic pyloric stenosis are diagnosed after a complete history and physical have been performed augmented by serum electrolyte evaluation, abdominal ultrasound, or upper gastrointestinal barium series. However, some patients can have equivocal radiographic testing and normal serum electrolytes. Two patients are presented in which esophagastroduodenoscopy was used to diagnose previously unsuspected infantile hypertrophic pyloric stenosis.

Health-related quality of life in pediatric patients with irritable bowel syndrome: a comparative analysis.

The primary aim of the study was to investigate the generic health-related quality of life (HRQOL) of pediatric patients meeting Rome II criteria for irritable bowel syndrome (IBS) in comparison to healthy children. The secondary aim was to compare pediatric patients with IBS to pediatric patients with Rome II criteria diagnosed functional abdominal pain (FAP) and patients with diagnosed organic gastrointestinal (GI) disorders. The study also investigated the associations between GI symptoms with generic HRQOL and evaluated group differences in school days missed and days sick in bed and needing care. HRQOL was measured with the PedsQLtrade mark 4.0 Generic Core Scales (Physical, Emotional, Social, and School Functioning). The PedsQLtrade mark was administered in outpatient pediatric gastroenterology clinics in California, Texas, and New Jersey. A total of 287 families (280 child self-report, 286 parent proxy-report) with children diagnosed with IBS (n = 123), FAP (n = 82), or organic GI disorders (n = 82) participated. Pediatric patients with IBS demonstrated significantly lower Physical, Emotional, Social, and School Functioning in comparison to healthy children, and comparable HRQOL to patients with FAP and organic GI diagnoses. GI symptoms were significantly associated with generic HRQOL. Patients with IBS demonstrated a significantly greater number of days missed from school, days sick in bed/too ill to play, and days needing someone to care for them than healthy children, but significantly fewer days than patients with FAP and organic GI disorders. Pediatric patients with IBS demonstrated impaired HRQOL in dimensions measuring Physical, Emotional, Social, and School Functioning. These findings suggest the need for targeted interventions to address these dimensions of impaired HRQOL.