Dollars and Jens
Thursday, December 03, 2009
Initial Claims of Unemployment
Today's number came in at 457,000. If my rubric is to be taken seriously, job losses have significantly slowed down in the past few weeks, and the prediction for next week is lowered to 430,000, and possibly 410,000–420,000 if we take the recent trend seriously.
If my rubric isn't to be taken seriously, you should probably just look at the seasonally adjusted data:
Labels: Initial Claims of Unemployment
Wednesday, December 02, 2009
Initial Claims of Unemployment
My (atheoretic) model's predictions, based on the last two weeks of information, are that the number tomorrow will be around 490,000, and then next week will drop into the low 450s. Yahoo has forecasts of 480,000 and 500,000 tomorrow, so I'm not contributing much.
I'm also predicting that, two weeks from now, I'll be less busy.
Labels: Initial Claims of Unemployment
Thursday, November 12, 2009
Initial Claims of Unemployment
502,000. Next Thursday expect 515000-520000, but something in the 490,000 range the day before Thanksgiving.
Incidentally, continuing claims are down on both an adjusted and unadjusted basis over the last few months, though this may be in some part people who ran out of benefits. I'm not sure that number contains a lot of useful information that isn't better represented in the monthly employment reports.Labels: Initial Claims of Unemployment
Friday, November 06, 2009
Initial Claims
The last prediction I gave, while not terrible, wasn't great, either, but failure doesn't deter a prophet. The report this morning was 512,000, and I'm predicting a dip a bit below 500,000 for next week, but a bounce back to 515,000 or so the week after that.
Labels: Initial Claims of Unemployment
Thursday, October 22, 2009
Initial Claims
The number came in at 531,000, which I think is fair "neener, neener" territory.
Next week, with no trend, the guess is about 521,000. The trend does look weaker in the last month or two than before, but I'd still think your best guess is a bit below 520,000, but probably not by more than the margin of error; let's say 515,000-520,000.
Labels: Initial Claims of Unemployment
Wednesday, October 21, 2009
predicted initial claims
My algorithm predicts that tomorrow morning's number will be 533,000. This is higher than the predictions Yahoo has, but it should be noted that it discounts the existence of a trend, and the trend is pointing downward. Since the trend has been about a weekly decrease of 5500, a best guess might be in the 525000 to 530000 area, with the understanding that there's some uncertainty here that probably exceeds 5500 anyway.
The main point I want to emphasize, though, is that, even accounting for a downtrend, the number is likely to blip up from last week. Don't worry; it will start dropping again after that.
Labels: Initial Claims of Unemployment
Thursday, October 08, 2009
Initial Jobless Claims
In my own parochial little world, the big news is that last week's number was revised slightly up such that the partially adjusted figure slightly exceeds the three week exponential moving average. Today's number is much lower, though; 521,000 adjusted, 502,000 partially adjusted.
Incidentally, the partially adjusted ARIMA(0,1,2) model would predict an adjusted number of about 507,000 next week.
Labels: Initial Claims of Unemployment
Thursday, October 01, 2009
Initial claims
A little bit of a deterioration. Yet another week below the 3-week exponential moving average for the partially unadjusted figure, but just barely.
Labels: Initial Claims of Unemployment
Thursday, September 24, 2009
jobless claims
I do, strictly speaking, have better things to be doing. The labor market continues on a trajectory of improvement that, at least by some measures, it has sustained consistently for half a year.
The (seasonally adjusted) jobless claims number from last week is 530,000, while the week before was revised from 545,000 to 552,000.
The endpoints on the lines were at 551,500 and 542,300, respectively.
Backing out some of the seasonal adjustment, the number is down three weeks in a row, to 504,000 from last week's 512,000.
These lines end at 515,000 and 523,000, respectively.
Labels: Initial Claims of Unemployment
Thursday, September 10, 2009
Initial Claims
Last week's number was revised up to 576,000, and this week's report is 550,000.
The smoothed number is 560,000.
Pursuing the idea that the seasonal adjustment has been overdone in a time when some of the seasonality has been wrung out of the labor market, last week's partially-unadjusted number is revised to 544,000, while this week's is 526,000.
That smoothes to 533333, where insignificant figures are included because of an aesthetic coincidence. The 3-week exponential moving average is down to 540,000:
The last time this partially unadjusted number exceeded its 3-week moving average was the report of June 18. (This is also true for the other smoothing mechanism, though that's had a few much closer calls.)
Labels: Initial Claims of Unemployment
Thursday, September 03, 2009
Initial Claims
The latest number is the same number as reported last week (570,000), though last week's number has been revised up slightly; in any case, it doesn't change the plots I've been producing much, so instead I'm going to put up plots based on partially-unadjusted numbers. In what follows I've taken the seasonally adjusted new claims and multiplied back in the fourth root of the seasonal factor.
Labels: Initial Claims of Unemployment
Thursday, August 27, 2009
initial claims
Let's see whether I still remember how to create and post plots.
Oh, good. And this one:
Initial claims reported this morning at 570,000, which is pretty close to where the series smoothes to, as well.
There was some appearance that seasonal factors were overadjusting the data; in particular, that low outlier in the adjusted series was a high outlier in the unadjusted series.
This week's adjustment factor (for data released next Thursday) is about the same as last week's, but then it bounces around a little bit near Labor Day; we may see a rise in this number two weeks from now followed by a dip due to the seasonal adjustment not being calibrated to current economic conditions.
Update:
If I back out a quarter of the seasonal adjustment, this is what the last 17 weeks look like:
Labels: Initial Claims of Unemployment
Thursday, July 30, 2009
Initial Claims
As has been the pattern, the seasonal adjustment has swung the change in claims in the opposite direction of the raw number; the seasonally adjusted number is another 30,000 higher than what was reported last week, with last week's number revised up as well, while the unadjusted number is down 78,000. The seasonally adjusted number is 584,000. The smoothed number is 575,000.
Seasonal adjustments for this data set tend to be largest in the summer; this will settle down when we're past the usual period of industrial furloughs and students working summer jobs. I'm traveling for a while; perhaps the data will be tame by the time I get back.
Labels: Initial Claims of Unemployment
Thursday, July 23, 2009
Initial Claims
The (-0,2,-0,2) smoothed line has jumped from 556000 to 562000. (The exponential moving average is down to 569,000.) The seasonally adjusted number from last week was revised up by 2,000 to 524,000, and the new number is up 30,000 to 554,000. The four-week sliding-window average is 566,000.
The unadjusted number is down 90,000 from the week before, rather the reverse of last week; both adjusted and unadjusted numbers are down about 10,000–15,000 from two weeks ago, as the seasonal adjustment has overadjusted in both directions in the meantime.
Labels: Initial Claims of Unemployment
Friday, July 17, 2009
seasonal adjustments and the business cycle
Note some important caveats about seasonally adjusted data:
On an unadjusted basis, the initial claims data showed a fairly large increase last weekâup 86,000 workers. But claims for unemployment compensation typically rise in early July as auto plants shut down to retool for the new model year. The jump in claims this July hasn't been as large as in years past since many of the auto plants were waylaid earlier in the year. So on a "seasonally adjusted" basis, the data showed a drop in claims of 47,000 workers.Fewer remaining factory workers might reduce the number of workers available to furlough. Similarly, some of the price increases pushing up the recent CPI report were actually decreases which might have run into support as already squeezed margins simply couldn't squeeze as much as usual.
Labels: Initial Claims of Unemployment
Initial Claims, yet again
As I expect this interests me more than anyone else, I'll try to restrain myself a bit here, but this line is the long-run (two weeks out) prediction that would be made based on data up to the time for which it's plotted using an ARIMA(0,1,2) model with both MA coefficients set to -0.2. It clearly hugs the data fairly well over any period of more than a couple weeks, it's previsible at the time of the data release to which you would compare it, and the change from one period to the next exhibits one third of the variance of the original series.
The last point on the new curve is 555,000, down from 590,000 the week before; the last two points have been significantly below recent values, and this figure responds more quickly to repeated deviations like that than the exponential moving average I posted before, while still filtering out two-thirds of the noise. I might, however, persist in using the exponential moving average just because 1) it's simpler to explain, and 2) it requires less "state"; if I give you today's exponential moving average, and you see next week's release, it's easy to calculate next week's average, while this requires two data points and is less mathematically simple. I haven't decided yet which I prefer, and I'd welcome any input anyone else wants to offer.
Labels: Initial Claims of Unemployment
Thursday, July 16, 2009
Initial Claims -- general notes
Incidentally, let me give some love to Karl Smith, whose posts on initial claims inspired me to start paying attention to (and posting on) a series I followed religiously for several years until about a year and a half ago.
He, like most of the world, smoothes the series with a 4-week sliding-window average; I use a three-week exponential, which I discovered a few years back frequently gave a better prediction of the next number than the consensus forecast did, except in the immediate post-Katrina period. I dislike sliding-window averages in particular for anything in which I'm looking for a trend; "trend" suggests I'm looking at changes from one week to the next, and the change from one week to the next with a sliding window is just a multiple of the change from one week to the week four weeks later, so it's noisy and is affected by a fair amount of old information. The change in an exponential moving average is more plausibly the innovation in that moving average.
It happens that the exponential-moving average amounts to modeling the series as an ARIMA(0,1,1). If you don't know what ARIMA is, it doesn't really matter; I didn't know what ARIMA was when I started doing this, and didn't actually work out that that's what this was until about a week ago; playing around in the last week, I find as a fairly robust result that ARIMA(0,1,2) seems to fit the data better, though the coefficients change a fair amount if I try fitting different 2-year data samples to that model. (ARIMA(0,1,1) seems to bounce around a bit, too, though a coefficient corresponding to a 3-week average doesn't seem to get much more than one standard error from the estimate regardless of the sample I look at.) I may start using a smoothing technique based on an ARIMA(0,1,2) at some point.
Labels: Initial Claims of Unemployment
Initial Claims of Unemployment
Last week's figure was revised up very slightly, but this week's number was 522,000, dropping the moving average to 576,000. This looks very good, at least by recent standards
Labels: Initial Claims of Unemployment
Thursday, July 09, 2009
initial claims
Seasonally adjusted initial claims was 565,000 last week. Note possible holiday effect and imperfect "seasonal adjustment". Still, good news.
Moving three-week exponential average:
High by the standard of an average point in the business cycle, but looks like a confirmed downtrend. The calendar issue makes next week's release more interesting.
Labels: Initial Claims of Unemployment
