Structural change detection in ordinal time series

Change-point detection in health care data has recently obtained considerable attention due to the increased availability of complex data in real-time. In many applications, the observed data is an ordinal time series. Two kinds of test statistics are proposed to detect the structural change of cumulative logistic regression model, which is often used in applications for the analysis of ordinal time series. One is the standardized efficient score vector, the other one is the quadratic form of the efficient score vector with a weight function. Under the null hypothesis, we derive the asymptotic distribution of the two test statistics, and prove the consistency under the alternative hypothesis. We also study the consistency of the change-point estimator, and a binary segmentation procedure is suggested for estimating the locations of possible multiple change-points. Simulation results show that the former statistic performs better when the change-point occurs at the centre of the data, but the latter is preferable when the change-point occurs at the beginning or end of the data. Furthermore, the former statistic could find the reason for rejecting the null hypothesis. Finally, we apply the two test statistics to a group of sleep data, the results show that there exists a structural change in the data.

The method of envelopes to concisely calculate semiparametric efficient scores under parametric restrictions

When addressing semiparametric problems with parametric restrictions (assumptions on the distribution), the efficient score (ES) of a parameter is often important for generating useful estimates. However, usual derivation of ES, although conceptually simple, is often lengthy and with many steps that do not help in understanding why its final form arises. This drawback often casts onto semiparametric estimation a mantle that can turn away otherwise able doctoral students or researchers. Here we show that many ESs can be obtained as a one-step derivation after we characterize those features (envelopes) of the unrestricted problem that are constrained in the restricted problem. We demonstrate our arguments in three problems with known ES but whose usual derivations are lengthy. We show that the envelope-based derivation is dramatically explanatory and compact, needing essentially two lines where the standard approach needs 10 or more pages. This suggests that the envelope method can add useful intuition and exegesis to both teaching and research of semiparametric estimation.

Approximate Learning of High Dimensional Bayesian Network Structures via Pruning of Candidate Parent Sets

Score-based algorithms that learn Bayesian Network (BN) structures provide solutions ranging from different levels of approximate learning to exact learning. Approximate solutions exist because exact learning is generally not applicable to networks of moderate or higher complexity. In general, approximate solutions tend to sacrifice accuracy for speed, where the aim is to minimise the loss in accuracy and maximise the gain in speed. While some approximate algorithms are optimised to handle thousands of variables, these algorithms may still be unable to learn such high dimensional structures. Some of the most efficient score-based algorithms cast the structure learning problem as a combinatorial optimisation of candidate parent sets. This paper explores a strategy towards pruning the size of candidate parent sets, and which could form part of existing score-based algorithms as an additional pruning phase aimed at high dimensionality problems. The results illustrate how different levels of pruning affect the learning speed relative to the loss in accuracy in terms of model fitting, and show that aggressive pruning may be required to produce approximate solutions for high complexity problems.

Evaluation of TOPS score as predictor for outcome in sick newborns

Background: Multiple parameters have been developed to prognosticate the outcomes of critically ill newborns admitted in NICUs. The objective of this study is to predict the outcome of newborns admitted in NICU using a simple but efficient score, TOPS score, involving alteration of physiological parameters. Aim of this study was to evaluate role of TOPS score in predicting mortality in sick neonates.Methods: The variables assessed under TOPS score on arrival for all subjects were: Temperature, Oxygen Saturation, Perfusion and blood glucose reading <45 mg/dl. All affected neonates were given treatment as per NICU protocol and outcome was assessed in terms of mortality or discharge using TOPS score. It was prospective study conducted at NICU, Department of Pediatrics, GMERS medical college and general hospital, Gotri, Vadodara. Study population was all admitted neonates aged <28 days at NICU.Results: Mean age of presentation of all cases was 2.8±3.58 days. Hypothermia on admission was observed in 63.3% cases. 40.8% cases had hypoxia. 26.5% neonates recorded poor perfusion. Mortality observed in hypoxic group was 51.7% followed by hypothermic group (46.9%). Highest strength of association was found for poor perfusion, mortality (87.5%) and OR-33.406. TOPS score was observed to be statistically significant (X2 value is 63.27, p < 0.05) as predictor of mortality. Thus, mortality rate increased with increasing no. of altered TOPS parameters. Regression analysis showed three factors (hypothermia, hypoxia, prolonged CRT) which are consistently associated with p value ≤ 0.05 for each variable and can be used to predict mortality.Conclusions: All parameters in TOPS score are physiologically important and each parameter carries an independent risk associated with mortality. It is important to note that multiple parameters affected increases the risk. TOPS score is a simple, basic and effective tool to guide about the condition of new born at admission and outcome. of neonatal mortality.

REGIONAL GOVERNMENT SPENDING EFFICIENCY ON HEALTH AND EDUCATION IN DECENTRALIZATION ERA : EVIDENCE FROM INDONESIA

The purpose of this paper is to assess spending efficiency of regional governments in Indonesia on health and education during the fiscal decentralization period year of 2010-2017. Relying on a sample of 33 provinces as regional government, this paper compute efficiency scores adopting nonparametric frontier that estimated by Data Envelopment Analysis (DEA) to study spending inefficiency. Results of the paper show that in west regions, Bali, Bangka Belitung, DI Yogyakarta, Jawa Tengah, and Kep. Riau relatively most efficient in public spending both on health and education in period of study. DKI Jakarta and Jawa Barat have efficient score on health, and Bengkulu has efficient score on education. On the other hand, in east regions, Gorontalo, Kalimantan Tengah, Kalimantan Timur and Sulawesi Utara were also most efficient in public spending on health and education services. Maluku and Sulawesi Tenggara have efficient score on health, and Kalimantan Selatan, Maluku Utara, Nusa Tenggara Barat, and Sulawesi Barat have efficient score on education. The results show that provinces in east regions of Indonesia were relatively more efficient in public spending both on health and education for promoting equal distribution of income

Data driven efficient score tests for Poissonity

New data driven score tests for testing goodness of fit of the Poisson distribution are proposed. They are direct applications of the general construction of data driven goodness-of-fit tests for composite hypotheses developed in Inglot et al. 1997. By a simulation study it is shown that these tests perform almost equally well as the best known solutions for standard alternatives and outperform them for more difficult alternatives.