There are two issues which have been occupying my mind ever since the trades that I had suggested have gone wrong. Type I error( trade when there is no signal) , i.e trade when the series is non stationary. Typically this is relevant to unit root testing where the null hypothesis says the series is random. Is the spread really a result of cointegrated time series ? Each issue requires detailed understanding of time series concepts.

Hallie King is a girl born with a rare genetic disorder, TAR syndrome , but whose story is quite inspirational.

Testing Stationarity of a series is an activity which has to be carried out by any quant trader, irrespective of whatever the frequency of data that is being used to develop models. Hence the amount of literature on Unit Roots, is VERY HUGE. Here is an interesting and simple way to test it. This note is refreshingly different from the usual stuff that one comes across in Unit Roots. It uses F test as a filter before going to the traditional tests.

“Millions long for immortality who don’t know what to do with themselves on a rainy Sunday afternoon.” - Susan Ertz, Anger in the Sky

A CUNY lecture on Random Matrices and its application in asset allocation models. http://fora.tv/embedded_player