Angela Noufaily
|
|
Research Associate in Statistics -- The Open University
|
Home Research Teaching Presentations Conferences Contact
Research interests
Generalised linear models (GLMs):
This includes ANOVA, ANCOVA and time series models (part of my BSc and MSc). I
continue to work with GLMs while exploring the regression-based Farrington method
for outbreak detection of infectious diseases. This includes improving this approach
and introducing new methods for outbreak detection.
Distribution theory:
Skills developed by studying the generalised gamma distribution in depth in my PhD thesis.
Maximum likelihood estimation using different reparametrisations and asymptotics were
developed. As part of my PhD thesis, inspired by the generalised gamma, I developed a new
distribution with interesting properties.
Quantile regression
Parametric quantile regression formed one of the main topics of my PhD thesis. The
quantiles of a generalised gamma distribution were fitted to data sets and compared
with other models. One of the main advantages is that the quantiles did not cross
(as is the case in the nonparametric cases) and they covered a wide range of shapes
that can be used for reference charts.
Survival analysis:
I am interested in modelling frailties using three- and four-parameter distributions
for explaining heterogeneity in transmission of infectious diseases. Using flexible
parametric models in survival analysis is a topic I plan to studydeeply.
Publications
6. Angela Noufaily, Yonas G. Weldesselassie, Doyo G. Enki, Paddy Farrington, Paul
Garthwaite, Nick Andrews and André Charlett (2013). Modeling Reporting Delays for
Outbreak Detection in Infectious Disease Data. (Awaiting decision from journal's
editor following the submission of first revisions).
5. Doyo G. Enki, Angela Noufaily and Paddy Farrington. A Time-Varying Shared Frailty
Model with Application to Infectious Diseases. (Accepted for Annals of Applied
Statistics). Open University Statistics Group Technical Report 13/01 available from
statistics.open.ac.uk.
4. Angela Noufaily and M. C. Jones. Parametric Quantile Regression Based on the
Generalised Gamma Distribution. (In Press: Journal of the Royal Statistical Society,
Series C). Open University Statistics Group Technical Report 12/02 available from
statistics.open.ac.uk.
3. Doyo G. Enki, Angela Noufaily, Paul H. Garthwaite, Nick J. Andrews, André Charlett
and C. Paddy Farrington (2013). Twenty Years of Automated Biosurveillance: Data From
the England and Wales Outbreak Detection System. Emerging Infectious Diseases, 19(1),
p. 35-42.
2. Angela Noufaily, Doyo G. Enki, Paddy Farrington, Paul Garthwaite, Nick Andrews
and André Charlett (2013). An Improved Algorithm for Outbreak Detection in Multiple
Surveillance Systems. Statistics in Medicine, 32(7), 1206-1222.
1. Angela Noufaily and M. C. Jones (2013). On Maximisation of the Likelihood for the
Generalised Gamma Distribution. Computational Statistics, 28, 505-517.