Data and prediction
Via Scott Sumner we saw the following article that mentions economic data and economic predictions. The statements that stood out to me were:
(Economic) predictions are, of course, the bread and butter of economic institutions. But can we believe them?
In recent years, some economists have begun to express doubts over predictions made from huge volumes of data, but they are in the minority. Most embrace the idea that more measurements mean better predictive abilities.
Hold up.
For one, as we have mentioned prediction is not the central element of what economists do – and even when they do predict the goal of such prediction is to give some view regarding risks and movements, not direct figures (it is more ordinal than cardinal in some sense).
Secondly, ever since the Lucas critique economists have been very nervous about predictions from large amounts of data without theory – I would say that the majority of economists doubt the usefulness of econometric models relying solely on huge amounts of data.
Economists would like data with less measurement error, that is closer to representing the true economic variables we discuss in theory – we aren’t looking for an infinite number of measures we can stick together to find a result. An economist that doesn’t use theory to inform their discussions of the economic outlook, but uses lots of data, isn’t an economist – that is all.
Good post. I just have a couple of questions.
How bad do you think the misunderstanding of economics is in the world, and where do you see it going over the next decade? How long do you think it will take before a proper understanding of economics permeates society? What real world results (i.e. on GDP growth, societal happiness, and the wages) do you expect from this broadening of understanding?
@Kimble
Hi Kimble,
I think the layman’s understanding of “social and economic issues” is actually pretty damned sophisticated – the main issue is that peoples view of how economists and other social scientists view the world is overtly simplistic. And to be fair, I guess often economists and other social scientists over simplify to explain things and leave themselves open to this interpretation.
The best we can hope is that, over time, the language of the discipline improves and becomes more transparent in society.
There is some talk that economists have an agency problem – they have to exaggerate the value of their expertise in order to earn extra returns. I don’t know how this works in other areas of the economics world, but its not something I’ve observed where I’m sitting.
As a result, like everything, I just hope that these discussion can be improved with “better information” – in this case I am sure that the physicists could have easily found this information, but they would prefer to leave the social sciences as an easy target for their critiques.
I have to disagree with your assertion that economists are nervous about predictions with large amounts of data without theory.
The Lucas critique teaches us that you have to be very careful about using estimated econometric relationships for policy making purposes, where changes to policy could have an effect on the estimated relationships. However, this does not preclude using large estimated models as a forecasting tool, assuming that policy will be set in the same manner as it has been in the past.
Indeed, there has been a huge literature over the past 10-15 years devoted to techniques for forecasting using large data panels, such as factor models, factor augmented VARs and large Bayesian VARs. These techniques generally give substantially better forecasting performance than models which are based more formally on economic theory, such as DSGEs.
@forecaster
Indeed, but one of the major criticism of VAR models in general is that it is difficult for us to figure out “why” things are happening.
Statisticians are happy to aim for some semblance of predictive accuracy as a primary goal, however economists add value through there ability to describe and explain.
Don’t get me wrong, forecasting and predictive accuracy are valuable useful things – but there needs to be a clear discussion of “why” things are happening in order to describe of explain.