Quantile estimation of semiparametric model with time-varying coefficients for panel count data

Panel count data frequently occurs in follow-up studies, such as medical research, social sciences, reliability studies, and tumorigenicity experiences. This type data has been extensively studied by various statistical models with time-invariant regression coefficients. However, the assumption of invariant coefficients may be violated in some reality, and the temporal covariate effects would be of great interest in research studies. This motivates us to consider a more flexible time-varying coefficient model. For statistical inference of the unknown functions, the quantile regression approach based on the B-spline approximation is developed. Asymptotic results on the convergence of the estimators are provided. Some simulation studies are presented to assess the finite-sample performance of the estimators. Finally, two applications of bladder cancer data and US flight delay data are analyzed by the proposed method.

Real Estate Bubbles and Contagion: Evidence from Selected European Countries

Using quarterly housing price-to-rent ratios from 1970 to 2018, this paper investigated the presence of real estate bubbles at a national level in eight selected European countries, namely Belgium, France, Germany, Italy, the Netherlands, Portugal, Spain, and the United Kingdom. Then, we analyzed bubbles contagion among these countries. We applied the generalized sup ADF test developed by Phillips et al. (2015) to detect explosive behavior in house prices. Subsequently, we implemented the non-parametric model with time varying coefficients developed by Greenaway-McGrevy and Phillips (2016) to estimate bubbles contagion among European real estate markets. We found evidence of at least one historical bubble in all these countries, with Germany, the Netherlands, Portugal, and Spain currently experiencing a rising bubble. The results also suggest that bubbles are contagious between these real estate markets.

The long-time behaviors of non-autonomous stochastic Nicholson’s blowflies systems with patch structure and time delays

In this paper, a class of non-autonomous stochastic Nicholson’s blowflies systems with patch structure and time delays is formulated and studied. By constructing suitable Lyapunov functions and using the stochastic technical, the pth moment boundedness and almost sure growth bounds are discussed, which reveal that solutions of the system do not exceed the time value [Formula: see text] and the sample Lyapunov exponent is no more than zero. Then, the system is proved to be exponentially stable (or extinct) if the production rate is less than the mortality rate, which provides an effective reference for the population control. Moreover, taking into account the specific form of time-varying coefficients, related results for several classical stochastic Nicholson’s blowflies systems are studied, and they show the significant improvement of this paper. Finally, numerical simulations for several specific examples are carried out to illustrate our theoretical conclusions.

1506Flexible parametric survival models investigating factors associated with diabetes-related foot ulcer time-to-healing

Background Diabetes-related foot ulcers (DFU) take months to heal, reduce patient’s quality-of-life, and induce large healthcare expenditure. Various factors have been identified to influence DFU healing at fixed periods, however, data on factors associated with time-to-healing is scarce. Methods Patients presenting with DFU to Diabetic Foot Services across Queensland, Australia between July 2011 and December 2017 were included and had their demographics, disease history and treatments examined at baseline. Outcome of interest was healing of all ulcers within two-year follow-up time. Time-to-healing and associated factors were examined using flexible parametric survival models, which easily enabled including time-varying coefficients and predicting proportions healed. Results Of 4,709 included patients (median age 63 years, 69.5% male, 10.5% Indigenous), median time-to-healing was 112 days, and 68% healed within two years. Younger age (<60 years), geographical remoteness, smoking, neuropathy, deep ulcers, infection, not receiving offloading, and no recent podiatry treatment were independently associated with longer time-to-healing. Time-varying effects of peripheral artery disease and ulcer size were identified for the first time: both had a negative influence on healing with effects diminishing after six months. The predicted proportions healed, for example, within six months is 65.0% (63.3-66.7) for people residing in a major city, 54.6% (52.6-56.8) in regional area, and 40.3% (34.6-47.1) in remote area. Conclusions This study identified novel and confirmatory factors influencing time-to-healing over 24 months in a large real-world cohort of people with diabetes-related foot ulcers. Visualizing the adjusted predicted proportion healed revealed the influence each factor had on healing rates over time. Key messages Flexible parametric survival model provided flexibility in investigating time-varying effects and outcome prediction in those with diabetes-related foot ulcer healing.