Linear filters used in seasonal adjustment (model-based or from the X-11 method) contain unit root factors in the form of differencing operators and seasonal summation operators. The extent to which the various filters (seasonal, seasonal adjustment, trend, and irregular) contain these unit root factors determines whether the filters reproduce or annihilate (i) fixed seasonal effects, and (ii) ...

With the great development of intelligent transportation systems (ITS), travel time prediction has attracted the interest of many researchers, and a large number of prediction methods have been developed. However, as an unavoidable topic, the predictability of travel time series is the basic premise for travel time prediction, which has received less attention than the methodology. Based on the...

In model-based software development, models are core development artifacts which are typically created and modified by multiple developers over a period of time. In order to be able to reason about the evolution of models, the computation of the differences between different versions of a model, called model differencing, is a crucial activity. However, in general a completely automatic approac...

Tracking information about fertilizers consumption in the world is very important since they are used to produce agriculture commodities. Brazil consumes a large amount of fertilizers due to its large-scale agriculture fields. Most of these fertilizers are currently imported. The analysis of consumption of major fertilizers, such as Nitrogen-Phosphorus-Potassium (NPK), Sulfur, Phosphate Rock, P...

We present a new method, median filtered differencing, for estimation of tangentdirection and curvature of digitised curves. On three synthetic examples and two images we show the algorithm performs successfully on both straight and curved segments even in the neighbourhood of discontinuities.

In a recent paper Jun Arima formalises the idea of preduc-tion { a non-monotonic schema providing a common basis for inductive and analogical reasoning. We examine some of the implications of this schema, as well as its connections with prototypes and conceptual spaces. A generalised version of his schema is motivated and developed to address diiculties with asymmetry and disjuncts.

This thesis investigates computational methods for pricing complex path-dependent derivative securities, especially geometricand arithmetic-moving-average-lookback options. The latter security was rst issued by Polaris Securities in 1999. Our methodology can be easily modi ed to price similarly structured options issued by other securities rms. The moving-average-lookback option is a call optio...

We introduce a methodology to predict when and where link additions/upgrades have to take place in an IP backbone network. Using SNMP statistics, collected continuously since 1999, we compute aggregate demand between any two adjacent PoPs and look at its evolution at time scales larger than one hour. We show that IP backbone traffic exhibits visible long term trends, strong periodicities, and v...

The training of precise speech recognition models depends on accurate segmentation of the phonemes in a training corpus. Segmentation is typically performed using HMMs, but recent speech recognition work suggests that the transient acoustic features characteristic of manner-class phoneme boundaries (landmarks) may be more precisely localized using acoustic classifiers specifically designed for ...