The finanscopics web site is devoted to understanding the statistical properties of financial time series, both for empirical data or for processes. The goal is to compare the statistical properties (the so called "stylized facts") in a systematic and uniform way, for various financial time series. This site presents many graphics visualizing various statistics. For example, the lagged correlation for the volatility (directly related to the heteroskedasticity) can be displayed for empirical FX time series (CHF/USD, JPY/USD, etc...) and for various processes (GARCH, long-memory ARCH, stochastic volatility, regime switching, etc...). For the return and the volatility, the main properties that are analyzed are the statistical distributions (with the possible fat-tails), various scaling for the distributions (related to the kurtosis), the lagged correlations for different vairables and in particular the heteroskedasticity, and some statistics related to the measure of time reversal invariance.

The background information and explanations about the statistics and graphs can be found in the following pages.

The plots for the various time series can be organized differently depending on the user goals. The following list presents the various ways along which the graphs are displayed and accessed.

The drop down menus on the upper right corner of each pages give you a quick access on the above pages.

This site offers a powerful mean to visualize and compare a large set of statistics, in a fairly raw form. The related book entitled "Discrete Time Series, Processes, and Applications in Finance" presents in details the motivations, definitions, and processes included in this site, the properties of the processes, as well as the conclusions that can be inferred from this study. Furthermore, the implications for important practical applications like market risk evaluation, option pricing or multivariate extensions are discussed in specific chapters.

The author of this site (Gilles Zumbach) can be reached directly by writing an email to the name 'gilles' at the present domain name ( Comments and feed-back (particularly when positive!) will be appreciated.