William T. M. Dunsmuir

VECTOR LINEAR TIME SERIES MODELS

This paper presents proofs of the strong law of large numbers and the central limit theorem for estimators of the parameters in quite general finite-parameter linear models for vector time series. The estimators are derived from a Gaussian likelihood (although Gaussianity is not assumed) and certain spectral approximations to this. An important example of finite-parameter models for multiple time series is the class of autoregressive moving-average (ARMA) models and a general treatment is given for this case. This includes a discussion of the problems associated with identification in such models. LINEAR PROCESSES; VECTOR ARMA MODELS; IDENTIFICATION; LIMIT THEOREMS;

Estimation of vector ARMAX models

The asymptotic properties of maximum likelihood estimates of a vector ARMAX system are considered under general conditions, relating to the nature of the exogenous variables and the innovation sequence and to the form of the parameterization of the rational transfer functions, from exogenous variables and innovations to the output vector. The exogenous variables are assumed to be such that the sample serial covariances converge to limits. The innovations are assumed to be martingale differences and to be nondeterministic in a fairly weak sense. Stronger conditions ensure that the asymptotic distribution of the estimates has the same covariance matrix as for Gaussian innovations but these stronger conditions are somewhat implausible. With each ARMAX structure may be associated an integer (the McMillan degree) and all structures for a given value of this integer may be topologised as an analytic manifold. Other parameterizations and topologisations of spaces of structures as analytic manifolds may also be considered and the presentation is sufficiently general to cover a wide range of these. Greater generality is also achieved by allowing for general forms of constraints.

Least Absolute Deviation Estimation for Regression with ARMA Errors

The asymptotic normality for least absolute deviation estimates of the parameters in a linear regression model with autoregressive moving average errors is established under very general conditions. The method of proof is based on a functional limit theorem for the LAD objective function.

Control of inventories with intermittent demand

Abstract The problem of controlling inventories with intermittent demands is considered. A method for determining re-order levels consistent with a specified customer service level is proposed. The distinguishing feature is the use of a probability distribution with a spike at zero to represent the relative frequency of periods with no transactions.

Estimation of Time Series Models in the Presence of Missing Data

Abstract A method is proposed for the estimation of models for discrete time series in the presence of missing data. Some justification is given for the use of this method over alternatives; the choice of estimator is likely to be governed by the pattern of missing data, the nature of the time series model, and computational considerations. The methods performance in estimating simple models is studied by simulations, and it is applied to a time series of pollution levels containing some missing observations.

Can spot urine protein/creatinine ratio replace 24 h urine protein in usual clinical nephrology?

Background:  This prospective study was designed to assess whether a single urine specimen, collected at the time of clinical assessment, could be used to estimate 24 h urinary protein excretion.

Maximum Likelihood Estimation for MA(1) Processes with a Root on or near the Unit Circle

This paper considers maximum likelihood estimation for the moving average parameter θ in an MA(1) model when θ is equal to or close to 1. A derivation of the limit distribution of the estimate θ LM , defined as the largest of the local maximizers of the likelihood, is given here for the first time. The theory presented covers, in a unified way, cases where the true parameter is strictly inside the unit circle as well as the noninvertible case where it is on the unit circle. The asymptotic distribution of the maximum likelihood estimator sub MLE is also described and shown to differ, but only slightly, from that of θ LM . Of practical significance is the fact that the asymptotic distribution for either estimate is surprisingly accurate even for small sample sizes and for values of the moving average parameter considerably far from the unit circle.

Can we predict recurrence of pre-eclampsia or gestational hypertension?

Objective  To estimate the rates of recurrence of pre‐eclampsia or gestational hypertension in a subsequent pregnancy and to determine factors predictive of recurrence.

Managing competing demands through task-switching and multitasking: a multi-setting observational study of 200 clinicians over 1000 hours

Objective To provide a detailed characterisation of clinicians’ work management strategies. Design 1002.3 h of observational data were derived from three previous studies conducted in a teaching hospital in Sydney, Australia, among emergency department (ED) doctors (n=40), ward doctors (n=57) and ward nurses (n=104). The rates of task-switching (pausing a task to handle an incoming task) and multitasking (adding a task in parallel to an existing task) were compared in each group. Random intercepts logistic regression was used to determine factors significantly associated with clinicians’ use of task-switching over multitasking and to quantify variation between individual clinicians. Results Task-switching rates were higher among ED doctors (6.0 per hour) than ward staff (2.2 and 1.8 per hour for doctors and nurses, respectively) and vice versa for multitasking rates (9.2 vs 17.3 and 14.1 per hour). Clinicians’ strategy use was significantly related to the nature and complexity of work and to the person they were working with. In some settings, time of day, day of the week or previous chosen strategy affected a clinicians strategy. Independent of these factors, there was significant variation between individual clinicians in their use of strategies in a given situation (ED doctors p=0.04, ward staff p=0.03). Conclusions Despite differences in factors associated with work management strategy use among ED doctors, ward doctors and ward nurses, clinicians in all settings appeared to prioritise certain types of tasks over others. Documentation was generally given low priority in all groups, while the arrival of direct care tasks tended to be treated with high priority. These findings suggest that considerations of safety may be implicit in task-switching and multitasking decisions. Although these strategies have been cast in a negative light, future research should consider their role in optimising competing quality and efficiency demands.

How comparable are road traffic crash cases in hospital admissions data and police records? An examination of data linkage rates

Objectives: An assessment of linked data was used to investigate the scope and the extent to which hospitalisations data and police crash records represent road crashes in New South Wales (NSW).