Indexing of Technical Change in Aggregated Data

My latest publication is forthcoming in Computational Economics and can be accessed here: The abstract:

The Baltagi–Grifn general index of technical change for panel data has earlier been applied to aggregated data via the use of period dummy variables. Period dummies force modeling into estimation of the latent level of technology through choice of dummy structure. Period dummies also do not exploit the full information set because the order of observations within periods is ignored. To resolve these problems, I suggest estimating the empirical equation for all possible structures of the dummy variables. The average over the different dummy coefcient estimates provides an index of technical change. More generally, the method estimates a general, model-free trend in linear models. I demonstrate the method with both simulated and real data.
The paper is essentially a real simple idea that works well in many situations and solves a difficult problem. I came up with this idea while working on this paper, and can as such be viewed as a spin-off from that. I gather that methodological papers seldom are spin-offs from empirical apply-this-different-method-to-this-data papers, but that is what happened here. I am quite proud of this paper and regard it as my most innovative and important contribution so far. What surprises me most is that, from what I can tell, no-one seems to have thought of this simple idea before.

Tags: , , ,

%d bloggers like this: