Customer Reviews

The Econometrics of Sequential Trade Models: Theory and Applications Using High Frequency Data (Lecture Notes in Economics and Mathematical Systems)

5 star:		(0)
4 star:		(1)
3 star:		(1)
2 star:		(0)
1 star:		(0)

 

 

The low rating is not about Dr Kokot's (solid) effort, but about Springer' deceptive marketing. The publishing house

a. dispenses with editors and proof-readers. (Memorably, the text has a comma before each connective "that", a grammar rule seen in Eastern European languages but not in English).

b. takes a research paper that, to a first approximation, presents one specific econometric model aligned to one specific theoretical model (Easley, Kiefer, O'Hara and Paperman, Journal of Finance, 1996) and publishes it as "Econometrics of sequential trade models", misleading readers about the book's scope.

c. goes with an $80 Amazon price tag for the result. ($109 on Springer's own site).

One paper, however great, is no substitute for a survey of a research area, and with this particular field being so dynamic, a 2004 time-stamp is a bit of problem. With many research papers available for free on SSRN and on university sites, I would hesitate paying even a third of the current price.

This book was very useful to me. It introduces the class of Sequential Trade Models (STM), letting the Easley, Kiefer, O'Hara and Paperman (EKOP) of 1996 play the center stage and include several expansions of it. To some degree I agree with an earlier reviewer that the book puts a lot of focus on EKOP and other models by the same authors, but I don't agree with the conclusion that this limits to book. The real value of the book comes from the chapter that ties STM with Finite Mixture Models (FMM) as with this framework you can expand on the class infinitely. EKOP and other models now become special cases. The drawback of the entire EKOP class of models is however that they depend on hidden variables (news/no news events and expansions of EKOP essentially widen this state space or the outcomes) so the markov switching events in the FMM that we estimate could actually be capturing a different state space than what the EKOP theory suggests and it would have been interesting to see to what degree the intraday model estimated in chapter 5 actually captures news/no news rather than some other variables, but it is not included.