… therefore, invoking the 5th amendment is sometimes a necessity.
a brief note by me about something interesting i read the other day, on the topic of godel’s theorem.
… therefore, invoking the 5th amendment is sometimes a necessity.
a brief note by me about something interesting i read the other day, on the topic of godel’s theorem.
interesting article by brennan and lo on how evolution may have driven some of the observed paradoxes of human behaviour, such as probability matching.
miller et al. present experimental evidence that human females have an estrous cycle, which shows up as a peak in lap dancers’ earnings synchronous with their menstrual cycles. interestingly, lap dancers using contraceptive pills showed no such earnings peak.
this paper by blume and stambaugh contains a warning that every quant should be aware of: price noise upwardly biases one’s expectation for total return, via jensen’s inequality.
bialek et al. show how to do clustering based on information-theoretic measures alone, which takes some of the ad-hoc-ness out of how folks did it previously.
raina et al. present a learning method which uses unlabelled data to help in forming a dictionary (sparse coding), that then helps in decomposing labelled data as in supervised learning.
lee & seung discuss how non-negative matrix factorization is good at finding features, in a way which is complementary to other techniques which encourage sparsity.
chigirev and bialek show how to apply the information bottleneck in a geometric setting:
title: “Optimal manifold representation of data: An information theoretic perspective”
abstract: “We introduce an information theoretic method for nonparametric, nonlinear dimensionality reduction, based on the infinite cluster limit of rate distortion theory. By constraining the information available to manifold coordinates, a natural probabilistic map emerges that assigns original data to corresponding points on a lower dimensional manifold. With only the information-distortion trade off as a parameter, our method determines the shape of the manifold, its dimensionality, the probabilistic map and the prior that provide optimal description of the data.”
a nice paper by pastor et al. shows that uncertainty in the growth rate of a company naturally leads to a higher stock price, due to convexity of the gordon growth model (a variant of discounted cash flows model). this idea can explain bubbles as being due to uncertainty in emerging technologies, such the one that developed during the internet bubble around year 2000.
it is well known that managers manipulate the books of their companies to meet performance targets. here is a paper by cohen et al., showing that that even warranty reserves are affected by this practice.