Seasonally adjusted data and the Anti-Christ

RobertBrusca's picture

About seasonal adjustment: the good, the bad and the ugly

Seasonal adjustment can ‘distort’ data. But, for the most part, if you know how it works, you can get some idea what sort of distortion certain circumstances might introduce. It is bad to ignore regular seasonal patterns when they occur because you can miss short terms shifts in trends that are important and that would more easily appear if data were adjusted. Economists use this form of data for a reason (hunker down for the criticism landslide here). If you ignore seasonally adjustment altogether you do not SOLVE most of the problems people blame seasonal adjusted for creating. In fact you introduce a whole new kind of error by making clumsy and stupid comparison. Seasonal adjusted is not perfect but it is good where the data go over the statistical hurdle for it to be applied. When it is workable, it is bad to ignore it or not use. To ignore it can produce some ugly results. Yes when you use seasonal adjustment the ‘seasonal factor’ kicks everything that happens whether it is seasonal or not and therein lies the potential for distortion. But unless you think we are being ‘shocked’ all the time, that is a small price to pay for a more stable and meaningful data-series.
Ignoring seasonal adjustment can be ugly- There are things that are 'seasonal' but that defy the calendar. For example, Easter is on the lunar calendar schedule not the Gregorian calendar. It pops up all over the Gregorian calendar more than Kurt Vonnegut’s Billy Pilgrim pops in out as he gets unstuck in time. If you were to compare the 'same' calendar week to the calendar week in the prior years, just previous to Easter, at Easter, and post Easter you would get distorted garbage.
Moreover, many of the people who rant about seasonally adjusted data claim that the weather is distorting the data today. The approach of looking at NSA data (Not Seasonally Adjusted) does NOTHING to alter that. Having a warmer, colder drier or wetter (spring, winter, summer or fall) is a factor to contend with whether data are seasonally adjusted or not. It is a problem that is independent of how the data are presented.
Also it’s leap year this year.
How do you account for that? What is the comparable 'week' for last year when there was no leap day? Seasonal adjustment takes account of that; NSA data do not.
When you RESTRICT yourself to 'NSA' data you can only look at Yr/Yr trends and you run the risk of missing the boat when trends turn. You cannot look at six month or even three month growth rates. And... picking the proper 'analog week' for last year is not nearly as straight forward as it seems...The week that lies 52 weeks ago may not be the best comparison…
There are 14 different 'types of years'. They can start with any day of the week (7) then each can be or not be, leap years (14). Then we can multiply that by noting the holidays that move around... to create many, many, more ‘types’ of years. Easter alone roams over a very wide range (look it up; there is a good internet site on Easter and how it aligns with the Gregorian calendar- it’s a mess!)

Seasonal adjustment programs have their problems; but we know what the methods are and can think about how the process might distort the data under certain circumstances.
Ignoring seasonal adjustment takes us back to the stone-age ( stoned-age ,dude?).
No thank you.
There is nothing highbrow or insightful about eschewing seasonally adjusted data. Zinging seasonally adjusted methodology because it is not perfect is throwing the baby out with the bathwater. Many people love to be critical of what they cannot - or refuse to- understand.
Don't be one of those. Of course if you do not LIKE or WANT TO ACCEPT the message in the ‘SA’ data, the NSA data may provide a refuge for the naysayer. And I suppose that is its true purpose more than people thinking seasonal adjustment is done by the anti-Christ or something.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
DeadFred's picture

It's not seasonal adjustments that we don't trust it's the BLS we very much don't trust. While I can't point to any evidence for fudged seasonal numbers (but how can I since it's is a black box?) there is ample evidence for biased data on the other numbers they throw out. Unemployment numbers are a great example. What percentage of times do the adjustments in the weekly new claims go in a market-unfriendly direction? It is not credible that their weekly adjustments are due to random data fluctuations. The best-case explanation is that they have a flawed analysis methodology that produces a systematic error which needs to be corrected each week. So why don't they fix the methodology? Maybe it's scheduled to be fixed sometime in 2018. The worse case is they are deliberately skewing the numbers in a market-friendly manner. So do we think they are incompetent or crooked? Neither gives us reason to trust their seasonally adjusted numbers. How many times have adjustments in the inflation algorithms made the inflation numbers go up rather than down. Why do they base their unemployment numbers on land-line surveys when the evicted, homeless and young aren't as likely to have land-lines? Could it be they resist changing methodologies that produce market-friendly errors while they are quick to adjust for iPads getting cheaper? We are being fed manure and kept in the dark because they want us to be happy mushrooms who buy lots of widgets and vote the incumbents back into office. Those of us who don't like the taste of cow dung have completely lost trust that the government numbers have integrity.

RobertBrusca's picture

Dear Fred, their methodology also includes face to face interviews.

I don't know how to construct an objective sample that includes vagrants.

The recent claims revisions have all been downward that is true. Its a hard-sell to bill that as a bias since they revise it the next week and the weekly revisions typically are chump change.



AldoHux_IV's picture

What is the methodology for the seasonally adjusted calculation? While I disagree with the bulk of the article and would like to see it for myself (as the fluff in the SA is what truly distorts data as well), what I find more alarming (and I could be dense) but the title included the anti-christ and I failed to see the connection in the content unless of course by pointing to a moving Easter you felt that it solved that literary device then it portends mostly to your lack in ability to properly outline the argument for blindly believing in seasonal adjustments. I myself would like to see a better data series to capture the true nature of the employment picture-- why aren't prisoners included?

RobertBrusca's picture

read the last line of the post

It's a snipe at those thinking there is evil behind all the government's doings -especially seasonal adjustment.

Did that really go over your head, shorty?.

Not the literary stretch you thought?

Argument is wholly logical - are you serious?.

and as to prisoners??!!
the Labor Force is always expressed as a the non-institutionalized population.. It excludes other institutionalized populations as well. Why?

They are not part of the market economy.

where would you put them?

(1) gainfully incarcerated?
(2) Minimum wage workers.
(3) Full time LIFETIME employment -just like JAPAN!!
(4) not unemployed and not in the labor force because they are not looking for work (just a way out?)

Bansters-in-my- feces's picture

Awk,Awk....Polly is crackers.

Bansters-in-my- feces's picture

Boy Robert,that was a lot of words to say nothing.

 Nothing would have made you look smarter.

RobertBrusca's picture

Ditto for you, whoever you are
hiding behind a made-up tag to snipe at people.
Good man!!

AcidRastaHead's picture

No problems with seasonal adjustments as a methodology.  It's when those indivdual inputs are essentially ignored as if they can't possibly provide insights I have a problem with.  More important is the need to explain/provide the methodology of the smoothing because without that disclosure it's nothing more than statistical alchemy.

hbjork1's picture

> statistical alchemy <

And, predictive statistical mathmatics are based upon the assumption of RANDOM DATA.  A die rolled or a coin tossed has no memory of what it did the last time or knowledge of what others are doing. 

As Merton, Scholes and Merriman found in the LTCM debacle that almost took down The Chase. Good Ole Alan Greenspan stepped up in the name of the people to bail them out. 

These stastical experts kinda forgot that their man is a communicating, remembering, thinking animal.

Plymster's picture

It is my understanding that the BLS and DOL seasonal adjustment methodology is a black box and has never been published.  How can anyone put any faith in a number that is arrived at with such a lack of transparency?  If you are aware of their seasonal adjustment methodology, can you post a link, or summarize it?  It's not that people don't understand seasonal adjustment, they just don't trust a hidden methodology, especially when the party hiding the methodology has something to gain by nudging the truth a bit.

Also, I'm not clear on the downside of looking at YoY values to form a trendline.  Is it because they only rely on one other datapoint?  Could you not build YoY values over a series of years to overcome this limitation?

RobertBrusca's picture

We know the general form of the process not the exact algorithm.

But it is not from Mars. We know how these process work. Seasonal factors are set in place before data are released so its not like someone tinkers with them to produce results.

There is no evidence of any data tinkering despite many claims. The allegations that have been the most public have been wrong and in some cases allegedly 'cooked' data turned out to have been under cooked and emerged as even stronger with the passage of time and further revision.

I'm not in the conspiracy camp. And I don't think we need to know everything about the governments models for seasonal adjustment to trust them. After all, economists from both parties work in the bureaucracy. That creates self-policing.

The problem with looking at only Yr/yr values is that it is harder to spot turning points and changes in momentum.

If you have ever worked with series that are not NSA you surely must have felt the constraints of their limitations. There are things you can do, but they are not as good.


fuu's picture

"I'm not in the conspiracy camp. And I don't think we need to know everything about the governments models for seasonal adjustment to trust them. After all, economists from both parties work in the bureaucracy. That creates self-policing."

There you go touting that "two party" meme. Keep talking, you out yourself with every word.

Lednbrass's picture

Yeah that line made me do a double take, that is naive to the point of childishness. Whatever PhD the man has, this is an infantile train of thought.

Im also a bit stunned that the author can argue for how great seasonally adjusted numbers are without knowing exactly what is involed in those adjustments, I cant decide if that is also naive or just plain stupid.  Without the ability to know and understand precisely what is behind the adjustments arguing that they have value is...well... I'll be kind and say perplexing. 

Nah, screw kind, its fucking stupid.  Without knowing what is weighted and how much any defense is based on nothing but blind trust and wishful thinking which is foolish.

Fox-Scully's picture

Its not seasonally adjusted data, it is politically adjusted data that occurs in all seasons!

Dingleberry's picture

Excuse me while I seasonally adjust myself....or is that hedonically?

RobertBrusca's picture

Hey, if works for you, its good for you.

Lee Adler- The Wall Street Examiner's picture

I take the opposing view. Actual data is better than massaged data, and it's not that difficult to analyze if you set up a spreadsheet. SA data randomly misrepresents reality. I mean, for goodness sakes, the DOL just restated 5 years worth of data based on newly derived SA factors. 5 years! 

Here's an example of how to get a truer pic using the actual numbers.

Why anyone would choose to look at obscure, massaged, fudge packed crap to the total exclusion of the actual numbers mystifies me. But that's what the entire mainstream punditry does. 


RobertBrusca's picture

I read your post post did you even read mine?

Just because data continue to be re-adjusted does not mean that all the re-adjustments were massive. The largest adjustments were for the current data where bench-marking and seasonal adjustment were involved.

On the contrary I have argue above that NSA data are a poor substitutes and with Easter closing in, your method is about run your people over the kind of bumpy road that could shake the head off of a bobble head doll mounted on the dash.

Good luck with that.

...and you still have not replied to the points I made above.

Lee Adler- The Wall Street Examiner's picture


No, I saw the headline and skimmed the first couple of paragraphs. That's all. 

It's nothing personal, but I never read or criticize other people's work. I just don't have time. I didn't read your earlier piece. In fact I've never read a single thing that you have ever written. The only people I read are Russ Winter and Bruce Krasting. They are my friends and colleagues, and I post their work, so I like to read it.

I really don't even have time for this response. But I saw the headline and thought that it contained a hidden, subtle, subliminal message that just possibly was aimed at me. So I posted a response to your headline.

The truth of the matter is that I'm just trying to get wider distribution of my work, hoping that a tiny fraction might find it interesting enough subscribe to my proprietary service. I do it so that I can continue to help support my mother, who has Alzheimers and is in assisted living, and whose life savings have been wiped out by Bernankecide, and so that I can continue to support my wonderful wife, whose business collapsed in the wake of the housing collapse. Our family income is much smaller than it was, and with my mother's situation, my overhead is much greater. I have to work really hard to keep everything afloat, and sharing my ideas here is just one of the many ways that I do that. 

I think that I can manage the data comparisons with a handful of moving holidays. The data goes back a long long time, and my spreadsheets cover all of it that's available. I have been a professional analyst for a very long time too. So yeah, I think I can handle Easter. And Passover, Rosh Hashanah, Ramadan, and Chinese New Year too. The warm weather can throw me though. I'm in Florida most of the time, and I sometimes forget to apply the warm weather adjustment.

I've been criticizing the way the mainstream media and economic establishment report economic data for a long time, and will continue to do so. And from time to time, when people take potshots at my research, maybe I'll respond. Controversy is good for my site traffic  

Meanwhile, readers can read my article and yours and decide for themselves which is more useful, if either. Most of the commenters here think I'm a jackass, so you have a good chance of winning that contest.





Fred Hayek's picture

So, if the author is going to acknowledge the variability of the seasons and the uncertainty of seasonal adjustments, should he also admit that applying seasonal adjustments to a LOSS of 2.7 million jobs and professing with confidence that we have *actually* gained some 200,000 jobs is ridiculous in that the scope of the uncertain adjustment dwarfs the size of the professed result?

RobertBrusca's picture

No Fred. That is the point.

When you regularly lose jobs in one month seasonal adjustment anticipates those losses and smooths over that by taking some job growth from a month that is regularly strong so you do not think 'normal' job loss in that month is a bad development it is a normal development.

Fred, I don't think you are even trying to understand what seasonal adjustment does.

DeadFred's picture

Robert you are seriously underestimating the sophistication of the average ZHer. We understand what seasonal adjustments do and even understand how valuable good adjusted data would be. What we don't have is any trust that these numbers are good data. The government has a vested interest in producing data that looks better than reality and we see numerous examples of skewed data. It makes all the government data suspect. ZH has posted articles pointing out the effects of the massive downturn in the first months of 2009 and how the adjustments may themselves need to be adjusted for that one-time event. I've taken years of stats classes and know full well how seriously one year's outlier data can bias adjustments. So the question for me is, do I believe the government has the integrity to adjust for those outlier data points or not? What I believe is more likely is that some lower level analysts like the CBO's Lan Pham tried to produce good numbers but their supervisors have bowed to political pressures and made sure the released numbers are 'Wall Street Approved'. Maybe I'm wrong, but you will get no traction with most of us by simply asserting that adjustments are made using good statistical methodologies. Try to convince us the numbers are honest, I would love to have reliable data to trust.

jayman21's picture

Most of us at ZH do understand what it does and that it creates problems for the sheep.  If things get to far out of whack, then we see social unrest.  Most of us here at ZH have a sense of community.  I am pretty sure you do too, or at least I hope you do.

Again, The problem I have is what do you do with it?  What is your next move?  Show me what we have missed in manipulating numbers to support a narrative.

As for forecasting, if you have not read Black Swan, I suggest you read it.  It is a good starting point.  I had to read it a few times to get it.  Mr. Taleb is a deep thinker and sometimes offensive.  His ideas are life changing.

While I am at it, also read "Confession of an Economic Hitman".  You will have a new view of trends in general.

mbasham's picture

ZH needs to incorporate an evolutionary survival of the fittest feature. Readers could either "rate" a contributor to death or alternatively opt to "ignore" a particularly stupid string of posts. 

Brusky, you may have degrees up the wazoo, or whatever, but your posts here are just stupid. You're simply aping the market, which is completing a bear market rally, leading you to view the world with spring rose colored glasses. I'll go as far as giving you the benefit of the doubt, you're simply a dullard trend following Cramer, which is perhaps all you can ever hope to be. Don't worry, you can be quite happy living a dull-tarded life.

RobertBrusca's picture

I do not follow Jim

I do not follow the 'crowd'.
Probably guilty on the other counts...

I think the consensus is how bad the economy is... You believe.
the Fed believes it
Obama believes
Republicans believe it.

I am optimistic and am in the minority

Stocks are going up not because those investors 'believe it' but because the bond market looks like Thin ice in the spring.

Good luck with that evolutionary survival.

drchris's picture

The economy is bad.  We may or may not be on the way to recovery, but nobody thinks the economy is good.  The stock market is pretty much back to normal, but GDP sucks, jobs suck and housing market sucks.  

jayman21's picture

Seasonal adjustments to see important trends or trend changes?  Are you a central planner?  Who the F*&^ cares if we let the market place move and reset as it should.

Someone help me out to understand why one would care?  I can come up with get reelected which will aid in herding the sheep.

jayman21's picture

Found this on him here at ZH:


Robert Brusca Ph.D. has been an economist on Wall Street since 1977. He has been a Division Chief at the NY Fed, a Fed-watcher at a major NY commercial bank and Chief Economist at major international securities firm. He is now an independent voice on the economy global trends and the political scene, operating a consulting firm in New York City.


Tucan sam knowns....he always knows the flavor of this case centrally planned fruit.

Kayman's picture

"an independent voice" ?

Another plant to try to convince us that the lying criminals are not lying criminals.

Seasonally adjusted data is little understood by the general public; that is why it is created.

Taking the lowest housing start months and then extrapolating them into yearly numbers is, at best, playing with numbers. At worst it is misleading. I go for the latter.


RobertBrusca's picture

yeah well I also dropped rear axles on trucks on the assembly lines in Detroit and had other menial jobs that may make me a better guru in your eyes.

I guess it is better to know nothing when you write about things?

Try that next time to go skydiving. I'll give you advice on how to pack a parachute.

johnQpublic's picture

36 bucks an hour dropping axels at an auto plant IS NOT menial

Schmuck Raker's picture

I can't help but suspect that owner, and "Chief" Economist Robert Brusca, Ph.D., of F.A.O. Economics* is the ONLY economist(or even employee) at the 'firm'.


* Fact And Opinion Economics

alexwest's picture

he is an idiot.. probably writing book about how great investor he is , so
need some publicity


RobertBrusca's picture

thanks for the value-added

YC2's picture

without getting into minutia.

sorry, on my phone...

YC2's picture

finally someone gets it!

it's a tool that must be understood in it's context.  Some data sets are seasonal.  Some recent years skew the adjustment.  You have to think through it just as with nsa data.  It is also a murky cover if shenanigans are going on.  Not just a concept to reject outright, tho.

I think YoY nsa comparisons would be the clearest shorthand way of looking at most data without

Dr. Engali's picture


"Ignoring seasonal adjustment takes us back to the stone-age "


I think the current path of lies and deceit inherent in  the system  are leading us back to the stone age.

RobertBrusca's picture

I think you have something there.

fuu's picture

Do you buy your hopium by the ton?

Rainman's picture

Seasonal adjustments are just another opportunity to flim-flam outcomes. More complexity means more  seemingly credible opportunities to arrive at desirable results.

Weil has a great analysis of corporate " earnings " presentations today. The truth on earnings is found somewhere in the muddy middle.

alexwest's picture

Don't be one of those. Of course if you do not LIKE or WANT TO ACCEPT the message in the ‘SA’ data, the NSA data may provide a refuge for the naysayer. And I suppose that is its true purpose more than people thinking seasonal adjustment is done by the anti-Christ or something.

only idiot could believe SA or NSA data from FED/NBER/BLS/NRA/etc..
doesnt matter. if SA data is compromised, then how can you believe NSA data..?

Jim Rogers once said " if you're in investment business and base you decisions
on GOV data, you are in wrong business"

before blubbering pointlessly this and that COULD WE SEE ANY TRADING RECORD?


therearetoomanyidiots's picture

Thanks for coming but it's time for you to go now. 

New American Revolution's picture

BULLSHIT!!!   Back in the 80's when I charted M-1 and M-2 money supply from the FED (VOLCKER), I only used the raw data sans the seasonal adjustment.   The season pattern showed up clearly on my charts so I could make my own adjustment, and the seasonal adjustments were most often missing because they were off a few days from prior years, but I charted the seasonal adjustments too just for grins, which only confirmed their meaninglessness.   You are far better off working with raw data as long as you have its history of repetition, which allows you to figure it out yourself.   The seasonally adjusted factors you keep around just so you can forecast how badly they will perform in distorting the raw data because, like I said, they are always off due to the subtle shifts in the calendar, which places say the 8th on Friday this year, but on Sunday the next year and so forth.   You see, the seasonally adjusted factors don't take this shift in to account and that is why they are pure wasteful bullshit.   Again, the only reason you look at that factor (and not the seasonally adjusted factor), is so you know how far off their seasonally adjust figures are going to be.   But it's all pure BULLSHIT.   Been there, done that.

RobertBrusca's picture

very interesting post.

Well I was a 'Fed watcher' in the 1980s and was employed to forecast weekly money supply numbers yes, seasonally adjusted numbers.

And did NOT forecast NSA numbers and I consistently had some of the very best forecast on the Street (I kept records) so I don't know what you were doing but the seasonal factors did not defy predictability they aided predictability.

At least for me it worked.

TimmyM's picture

Thanks for your contribution to this website. One-way dialogue is boring.


fuu's picture

"One-way dialogue is boring."

You can find everything Robert posted above hawked on the approved media outlets like CNBC, Bloomberg, MSNBC, ABC, CBS, NBC, FOX, etc.

I find it strange that people want to bring the real echo chamber into this forum all in the name of dialogue.

Kayman's picture

I think this is Tyler's version of red meat.

Come on now, Leo, it's you, isn't it.