Martin Sewell

Learn More
This paper provides an exhaustive review of the literature on the characterization of financial time series. A stylized fact is a term in economics used to refer to empirical findings that are so consistent across markets that they are accepted as truth. Financial time series may be characterized by the following stylized facts. The autocorrelation of(More)
This note presents a chronological review of the literature on ensemble learning which has accumulated over the past twenty years. The idea of ensemble learning is to employ multiple learners and combine their predictions. If we have a committee of M models with uncorrelated errors, simply by averaging them the average error of a model can be reduced by a(More)
The term kernel is derived from a word that can be traced back to c. 1000 and originally meant a seed (contained within a fruit) or the softer (usually edible) part contained within the hard shell of a nut or stone-fruit. The former meaning is now obsolete. It was first used in mathematics when it was defined for integral equations in which the kernel is(More)
The basic idea behind the Fisher kernel method is to train a (generative) hidden Markov model (HMM) on data to derive a Fisher kernel for a (discriminative) support vector machine (SVM). The Fisher kernel gives a ‘natural’ similarity measure that takes into account the underlying probability distribution. If each data item is a (possibly varying length)(More)
Many philosophies treat emotions as the polar opposite of logic, and an economist would judge emotions as irrational. Taking an evolutionary perspective informs us that emotion is no less rational than any other behaviour, and likely evolved to maximize lifetime utility. Frank (1988) argues that emotions evolved to enable us to sacrifice a short-term(More)
The idea of ensemble learning is to employ multiple learners and combine their predictions. There is no definitive taxonomy. Jain, Duin and Mao (2000) list eighteen classifier combination schemes; Witten and Frank (2000) detail four methods of combining multiple models: bagging, boosting, stacking and errorcorrecting output codes whilst Alpaydin (2004)(More)
The ‘selfish gene’ is an excellent metaphor. Evolution seeks to maximise gene replication within the local reproducing group in the context of the environment. It does so by distributing genes among vehicles (organisms) which are each motivated to maximize their inclusive fitness, making the most of the genes that they have been allocated. We show how this(More)
The problem of how to maximize growth of wealth was solved over half a century ago by Kelly and Breiman: maximize the expected value of the logarithm of wealth after each period. However, as living organisms we have evolved to maximize gene replication, not wealth. Natural selection is a slow process, so our minds today are adapted to seeking our ultimate(More)