Search results for " correlation matrix"
showing 4 items of 4 documents
Sparse model-based network inference using Gaussian graphical models
2010
We consider the problem of estimating a sparse dynamic Gaussian graphical model with L1 penalized maximum likelihood of structured precision matrix. The structure can consist of specific time dynamics, known presence or absence of links in the graphical model or equality constraints on the parameters. The model is defined on the basis of partial correlations, which results in a specific class precision matrices. A priori L1 penalized maximum likelihood estimation in this class is extremely difficult, because of the above mentioned constraints, the computational complexity of the L1 constraint on the side of the usual positive-definite constraint. The implementation is non-trivial, but we sh…
On the origin of power law tails in price fluctuations
2003
In a recent Nature paper, Gabaix et al. \cite{Gabaix03} presented a theory to explain the power law tail of price fluctuations. The main points of their theory are that volume fluctuations, which have a power law tail with exponent roughly -1.5, are modulated by the average market impact function, which describes the response of prices to transactions. They argue that the average market impact function follows a square root law, which gives power law tails for prices with exponent roughly -3. We demonstrate that the long-memory nature of order flow invalidates their statistical analysis of market impact, and present a more careful analysis that properly takes this into account. This makes i…
Collective Behavior of Price Changes of ERC-20 Tokens
2020
We analyze a network constructed from tokens developed on Ethereum platform. We collect a large data set of ERC-20 token prices; the total market capitalization of the token set is 50.2 billion (109) US dollars. The token set includes 541 tokens; each one of them has a market capitalization of 1 million US dollars or more. We construct and analyze the networks based on cross-correlation of tokens’ returns. We find that the degree distributions of the resulting graphs do not follow the power law degree distribution. We cannot find any hierarchical structures nor groupings of ERC-20 tokens in our analysis. peerReviewed
Spectral density of the correlation matrix of factor models: a random matrix theory approach.
2005
We studied the eigenvalue spectral density of the correlation matrix of factor models of multivariate time series. By making use of the random matrix theory, we analytically quantified the effect of statistical uncertainty on the spectral density due to the finiteness of the sample. We considered a broad range of models, ranging from one-factor models to hierarchical multifactor models.