Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: In a recent article, Wood (2000) exhorts researchers to "take time varying relationships seriously," yet stops short of adopting a Bayesian approach that has the necessary flexibility to do so. While the Bayesian analysis of time-serial data using tools such as the dynamic linear model (DLM) has a fairly long history in econometrics and applied statistics, there have as yet been no applications in political science. In this paper I briefly introduce the DLM and illustrate its use with some simple examples. I then turn to an examination of the influence of ideological think-tanks on public mood, using data from Smith's (2000) study.
Check author's web site for an updated version of the paper.

Publication Type: Abstract
Abstract: Rationale: Health care expenditure regressions are used in a wide variety of economic analyses including risk adjustment and program and treatment evaluations. Two recent articles have demonstrated that generalized gamma models with heteroskedasticity (GGM-het) and extended general linear models (EGLM) provide flexible approaches to deal with a variety of data problems commonly encountered in expenditure estimation. To date, however, there have been few empirical applications of these models to expenditures.

Objective: We use nationally representative data from the first six panels of the U.S. Medical Expenditure Panel Survey (MEPS) to compare the bias and predictive accuracy of GGM-het and EGLM models with other regression models in a cross-validation study design.

Methodology: We estimate models of prescription drug, ambulatory and total health care expenditures conditional on having any expenditure. Models are estimated separately for the elderly and other privately insured adults. Since expenditure distributions vary by type of service and population, the appropriate functional form is also likely to vary. In estimating expenditures, we focus on two recently developed modeling approaches that flexibly accommodate skewness, kurtosis, heteroskedasticity and other data problems. The GGM-het model, proposed by Manning, Basu, and Mullahy (2005), uses a log-link like many standard GLM models. However, the GGM-het model is more flexible than standard models because the generalized gamma distribution has a scale parameter and two shape parameters and variance is explicitly modeled as a function of explanatory variables. In the EGLM model, proposed by Basu and Rathouz (2005), the link function is not specified prior to estimation. Instead, both the link and variance functions are simultaneously estimated along with the coefficients.

Our models use socioeconomic characteristics and condition information from the first year of each MEPS panel to predict expenditures in the second year. We use a split-sample cross validation design to compare results from GGM-het, EGLM, log OLS with heteroskedasticity (log-het), linear OLS, Poisson and Gamma models. We use the validation sample to test for over-fitting and to examine predictive ratios and mean prediction errors in the entire sample, in the tails of the distribution and for persons with chronic conditions.

Results: In our preliminary analysis we focused on total expenditures and estimated all types of models except EGLM. We found that the expenditure distribution for the elderly was more kurtotic than the distribution for other adults and the distributions varied in the extent of heteroskedasticity beyond simple functions of the mean. Overall, the GGM-het and log-het models fit the data for privately insured adults very well. However, none of our models was clearly superior for the elderly.

Conclusions: Our preliminary analysis confirms that GGM-het models are robust to a wide variety of common data problems. For some distributions, however, an even more flexible estimator, such as the EGLM model, may be required.

Publication Type: Conference Paper/Unpublished Manuscript
Abstract: This paper looks at the War on Terror through concepts drawn from the theories of the new scientific paradigm offered by complexity. Jihadist networks, such as those behind the September 11 attacks and the bombings in London and Madrid, are discussed in terms of complex adaptive systems, emergent organisations that coalesce and self-organise in a decentralized fashion. Such an analysis breaks with the misleading depiction of al-Qaeda as a monolithic organisation directing global operations from the top and allows for a greater understanding of the adaptability and resilience of jihadist networks in the face of strenuous state efforts to identify and eradicate them. Complexity can shed new light on the War of Terror by providing an account of the bottom-up self-organisation of jihadist networks, the social dynamics of radicalisation which bring new individuals into them, and the systems of distributed intelligence which allow them to operate and pursue successful attacks on the basis of partial and localised information.

Publication Type: Conference Paper/Unpublished Manuscript
Abstract: As a mathematician and violinist, I have always been intrigued by
patterns in melody lines. No one can dispute the themes that
various composers such as Beethoven and Brahms place throughout their
symphonies. In mathematics, linear recurrence relations provide a means
of generating sequences based on previous elements of the sequence.
Inspired by a desire to quantify certain melodies, this talk will include
a discussion of linear recurrence relations
in music. In particular, whether or not we can find exact or approximate
linear recurrence relations in classical works, independent of
the key signature of a piece. We will also use
such linear recurrence relations to generate melodic sequences and see how
closely they resemble the original work. This also provides a way
to introduce linear recurrence relations to students by
listening to the relationships. Analyses are performed in
Matlab using standard Musical Instrument Digital Interface (MIDI)
file format. We conclude by letting the audience ``compose" music based
on a recurrence relation of their choice!

Publication Type: Conference Paper/Unpublished Manuscript
Abstract: One of the problems teaching a linear algebra class is that the students are always complaining about how they don’t see any use for the material. To address this problem I assign semester projects to pairs of students where each pair will examine a different application of linear algebra. The students will study this application, create some examples of their own illustrating this use of their topic, write a 4 to 7 page paper on this topic and do a classroom presentation of their work. In this talk I will look at these assigned applications and my expectations for the project, in addition to my assessment of the value of these projects in linear algebra.