I am taking a course in Econometrics, although I am not sure I will be using that method in my PhD thesis. For the uninitiated, econometrics is the "application of statistical and mathematical theories in economics for the purpose of testing hypotheses and forecasting future trends", according to Investopedia. It is basically trying to infer cause and effect from a given set of data. I have never been intellectually convinced about its efficacy or legitimacy, but so strong is the consensus in economics that any theory must be testable and falsifiable that you need to use it to be taken seriously. So, I did use it in my master's thesis and was complimented for my work by the examiners but I know what I did was bull-shit.
For a long time after that, I thought econometrics, especially of the time series variety, was bull-shit. I don't think so any more because I have now interacted with genuinely bright econometricians who seem to know what they are doing. They are able to answer my questions intelligently and if not all the answers are comprehensible to me, it may be due to my powers of comprehension, or lack thereof. Now, I do feel econometrics is a worthwhile endeavour but it may not be worthwhile for me. I will have to spend too much time understanding it properly to convince myself of its efficacy and, frankly, I think that time will be better spent elsewhere.
Of course, try explaining that to the statheads, two of whom inhabit my PhD room. They believe any theory in the world must be statistically testable and that empirical (or more narrowly statistical) rigour is the only worthwhile form of intellectual rigour. To them you can say anything you want as long as you can prove it with statistics.
I have long believed that one can play with data until you find the result you are looking for. I don't think that's an easy exercise--you will be caught out by smart people. These people are well aware of the limitations of econometrics and the strong assumptions that undergrid it. You can dismiss econometrics by saying that you don't believe in the assumptions, just like many non-economists dismiss neoclassical economics by saying that the assumptions are unrealistic. But such a position often becomes an excuse for not engaging with a difficult topic more closely. And, with the hegemony that econometrics enjoys, the detractors who don't really understand it can be easily dismissed. The intelligent, open-minded econometricians I have come across are far less dismissive of social theories, but they are few and far between. Economics does attract a disproportionate share of statheads and mathheads who are downright dismissive of any research that doesn't include statistics.
Perhaps it is healthy to be around people who think your research is bull-shit because it forces you to reflect more closely on your methods. Sometimes, it can weigh on your confidence, but if you are prone to over-confidence like me, maybe that is a good thing. But, at least I can say that I am more open to their research than they are to mine, even I can't be bothered to spend time understanding it. I suppose my philosophy is: "I know you are smart as a result of our interactions or because of your successes or because people who I think are smart think you are smart. Which is why I think what you are doing is worthwhile even if I don't have the time or intellectual capacity to understand it."
For a long time after that, I thought econometrics, especially of the time series variety, was bull-shit. I don't think so any more because I have now interacted with genuinely bright econometricians who seem to know what they are doing. They are able to answer my questions intelligently and if not all the answers are comprehensible to me, it may be due to my powers of comprehension, or lack thereof. Now, I do feel econometrics is a worthwhile endeavour but it may not be worthwhile for me. I will have to spend too much time understanding it properly to convince myself of its efficacy and, frankly, I think that time will be better spent elsewhere.
Of course, try explaining that to the statheads, two of whom inhabit my PhD room. They believe any theory in the world must be statistically testable and that empirical (or more narrowly statistical) rigour is the only worthwhile form of intellectual rigour. To them you can say anything you want as long as you can prove it with statistics.
I have long believed that one can play with data until you find the result you are looking for. I don't think that's an easy exercise--you will be caught out by smart people. These people are well aware of the limitations of econometrics and the strong assumptions that undergrid it. You can dismiss econometrics by saying that you don't believe in the assumptions, just like many non-economists dismiss neoclassical economics by saying that the assumptions are unrealistic. But such a position often becomes an excuse for not engaging with a difficult topic more closely. And, with the hegemony that econometrics enjoys, the detractors who don't really understand it can be easily dismissed. The intelligent, open-minded econometricians I have come across are far less dismissive of social theories, but they are few and far between. Economics does attract a disproportionate share of statheads and mathheads who are downright dismissive of any research that doesn't include statistics.
Perhaps it is healthy to be around people who think your research is bull-shit because it forces you to reflect more closely on your methods. Sometimes, it can weigh on your confidence, but if you are prone to over-confidence like me, maybe that is a good thing. But, at least I can say that I am more open to their research than they are to mine, even I can't be bothered to spend time understanding it. I suppose my philosophy is: "I know you are smart as a result of our interactions or because of your successes or because people who I think are smart think you are smart. Which is why I think what you are doing is worthwhile even if I don't have the time or intellectual capacity to understand it."
No comments:
Post a Comment