There’s news and there’s news

December 16, 2011

Two recent posts included the word “news”, but in different senses.

Events

“News” in the sense of reports on events was discussed in “News analytics”.  We can think of this as an approximation to objective reality.

Market moves

The post “Volatility estimation and time-adjusted returns” used “news” in the sense of “that which moves market prices”.  News in the first sense is only part of this sort of news.  This sort also includes market participants watching each other.

A surprise (to me at least) was how little effect non-trading days — weekends and holidays — seem to have on the size of returns for the days following them.

Self excitement

It occurs to me that perhaps there is a way of decomposing volatility into exogenous and endogenous components.  The exogenous part would be measured by the flow and sentiment from news analytics data, the endogenous would be the remainder.

Almost surely a stupid idea, but there it is.

Subscribe to the Portfolio Probe blog by Email

Latest posts

Leave a Reply

Related posts

  • April 21, 2011

    What is the effect on predicted and realized volatility of substituting risk fraction constraints for weight constraints? Previously This post depends on two previous blog posts: "Unproxying weight constraints" [...]

  • April 18, 2011

    How well do asset weight constraints constrain risk? The setup In "Unproxying weight constraints" I claimed that many constraints on asset weights are really a proxy for constraining risk. [...]

  • March 28, 2011

    Overfitting is a problem when trying to predict financial returns.  Perhaps you've heard that before.  Some simple examples should clarify what overfitting is -- and may surprise you. Polynomials [...]