Heavy Lifting - thoughts and web finds by an economist
|I also contribute to Division of Labour||Load HL's Front Page|
Friday, April 02, 2004
I'm not a forecaster but I play one on t.v.
Yesterday around noon UTA's media relations department invited us economists to make a forecast about today's labor market report. My first inclination in these cases is to dismiss them as frivolous and misapplications of economists' time and efforts. However, after a few moments of reflection I decided to take a "stab" at predicting unemployment - having no expertise in the area - and to use the opportunity to make a statement about factor productivity and why employment is a lagging indicator.
I then had to decide how to make my prediction. I didn't have time to set up a large sample of time series data on unemployment, estimate ARIMA models, perhaps a few Vector Autoregression models and look at impulse response functions, so I needed an alternative approach. I spent about twenty seconds deciding whether I could get away with the cynical prediction that unemployment would either go up, go down, or stay the same, but dismissed the idea. Finally, I decided to calculate the averages of unemployment and employment gains over the past decade (data from BLS) and use reversion to the mean as the decision rule. This led me to predict unemployment at 5.5% and job increases somewhere around 150,000.
Turned out I was wrong. Oh well, off by two percentage points in the unemployment figure and off about 150,000 in the new jobs area. Most people might think that that is a pretty good prediction (hell, weathermen keep their jobs after worse predictions), but in economics forcasting the mistake, especially in the jobs area, is intolerable. Therefore, this round my back of the envelope approach failed miserably this time. Good thing my prediction didn't have any money riding on it. Perhaps I will have better luck next time? After all, eventually the unemployment rate will once again be 5.5% at the same time that new jobs total 150,000 - right?
Anyway, below I post my "prediction" from yesterday.
Comments: Post a Comment
Le Chai - galerie du vin
Posts that contain Craig Depken per day for the last 90 days.
Heavy Lifting's Main Page
Money I Found Today
Heavy Lifting - Firehose style (56k warning)
Modified maystar design
powered by blogger