Shocks & Surprises & Stockmarkets

Let's entertain a proposition: if stockmarket prices move according to changes in information, and if some of the information which matters is economic information, it may be that there's a relationship between the economic shocks and surprises with which we are buffeted daily, and movements in stockmarkets.

This strikes me as reasonable and even plausible. (Though not, please note, a complete guide to stockmarket behaviour. The thought is only it may be one input amongst many.)

My Shocks & Surprises Weekly four-pager (email me if you want to take a look), contains an attempt to track the general trend of shocks & surprises. I take the net percentage of surprises minus shocks over the previous six week period, and then express that percentage as a number of standard deviations, assuming the errors from consensus and/or trend in the economic data I watch is normally distributed. So, for example, in the last six weeks I've checked out 405 pieces of economic data, of which 25.7% were positive surprises, and 17.3% were negative shocks. This means a net 8.4% were positive surprises, which works out at 0.52SDs of a normally distributed population.

If you're going to mess with data, one should at least do it in full public view. So I also used a six-week average for the S&P 500 and the MSCI World Free Index to see if there how the proposition might hold up. Here are the results:

No, I'm not suggesting that the proposition is proved, or that changes in the balance of Shocks & Surprises necessarily presage changes in stockmarket direction. Nothing is proved over a four-month period.

But it is interesting. I'll keep it under review.