Rocket science to some, a waste of a good mind to others - either way, modern econometrics is overdue a serious evaluation. Christopher L. Gilbert of the Università degli Studi di Trento and Duo Qin of Queen Mary, University of London give it a go in their recent Queen Mary working paper The First Fifty Years of Modern Econometrics. Here's the abstract:

We characterize modern econometrics in terms of the emergence a widely accepted analytical framework. A major theme which dominated much of the debate through the century was whether and how econometric models can reflect theory-generated economic structures. In the period prior to the 2nd world war, economists adopted a wide variety of analytical methods, some ad hoc but others reflecting advances in statistical methodology.

Business cycle analysis and demand analysis were the two major areas in which statistical theory was employed. Methods became increasingly formalized but problems of data adequacy, estimation and identification were not always well distinguished. During and immediately after the war, Cowles Commission research sought to base econometrics on autonomous probabilistic models specified in terms of underlying structural parameters. Least squares would not normally be consistent in such models and maximum likelihood estimation was to be preferred.

Subsequently, however, the pendulum swung back towards least squares-based methods and this was reflected in the textbook expositions of what was accepted as standard econometrics in the late sixties and early seventies. Subsequently, the paradigm was undermined by the challenges imposed by rational expectations modelling, which challenged standard identification assumptions, and by the poor forecasting performance of many macroeconomic models by comparison with black box time series competitors. The result was a revival of non-structural modelling, particularly in the analysis of macroeconomic data.

Quite dry, alas - but the paper itself is more readable (Adobe won't let me copy the text from the PDF).

Econometrics and fractal analysis share the commonality of a

attempting to construct or define a methodology to describe in

a purposeful and predictive way the macroeconomy. For the fractal

analysis small school perhaps ' a breakthrough' has occurred....

The fractal decay puzzle which retrospectively will epitomize

maximum mathematical efficiency in the three fractal primary decay

process must be solved prior to its evolution in order to validate

fractal analysis as a fundamental 'science' of macroeconomics.

At its smallest time units, equity valuation growth to trading

saturation points and decay to selling saturation points makes

intuitive sense. At the longest time units, growth to the consumer

debt saturation point based both on collective limited wages and

overvaluation and overproduction of assets which in turn are based

both on low interest rates and creative easy lending practices is

followed by a contraction and decay process of asset values that

unmasks the surpluses in the system and results in job contraction and

lower wages based on those ongoing surpluses and contracting prices.

This longer perspective likewise makes intuitive sense.

Equity valuation fractals quantify, in the most efficient manner, this

qualitative intuitively reasonable process. The markets are not - they

are not random walks. The markets are defined by dollars available

for investment, current total value of investments and assets, ongoing

wages, outstanding debt obligations, and inflationary pressures of day

to day living fueled by low interest rates, imprudent lending

practices, and inappropriate money creation in relation to traditional

debt to wage ratios. The numbers 1-10 and fractions and multiples

thereof define the ongoing valuations, debt, and money supply that is

the market. That the market 'marches' and otherwise evolves to a beat

of efficient fractal patterns that represents the total integration

of these numbers into repetitive maximum efficient patterns is wholly

reasonable.

The probability that the retrospectively identified recurrent fractals

patterns occur by chance alone approaches zero. The challenge remains

to decipher the non complex puzzle and to delineate the decay process

prior its real time evolution. If it can be done, it will be the first

time that quantification of the valuation decay process has been

accomplished before the event. Ludwig Von Mises and Joe Kennedy by

different methodologies in 1928 and 1929 intuitively sensed in a

qualitative way the lurking decay process and collapse and acted

accordingly.

Just as in 1929 and 1720, consumer saturation and/or asset

overvaluation relative to ongoing wages and debt load will be the

precipitating cause of the mechanistic deterministic non stochastic

fractal decay process. By lowering the fed fund rates to extreme low

levels and not controlling lending practices, the Federal Reserve and

government caused excess borrowing, excess valuation, and excess

production - with expected excess pain in the subsequent natural and

mechanistic devolution process. By its ongoing raising of fed fund

rates the Fed has placed boron rods into the uncontrolled reactor and

decreased the highs of overvaluation with an expected lessening of

pain in the devolution. Hence the Euro- Nikkei markets with

relatively less tightening by their central banks have had higher

recent terminal valuations. Regardless of the degree of

overvaluation, devolution will inevitably occur and will inevitably

occur on time according to the most efficient fractal decay

progression Just as the Federal governmen and the Federal Reserve

with its large check book and rapid lowering of interest rates,

respectively, could not change the couse of the devolution in

1929-1932 and

2000-2003, so will they be powerless to change the course of the

present devolution.

The perfect often stated fractal growth pattern is x/2.5x/2x followed

by an idealized decay of 1.5x. This last 1.5x can be subdivided into

an idealized decay fractal pattern of y/2.5y/2.5y.

TNX, the ten year note, had an exhaustion gap on Wednesday 26 October,

day 38 of a

19/48/38 of 38 perfect growth sequence. What is possibility that this

perfect x/2.5x/2x growth fractal with an exhaustion gap on the last

day of the third growth fractal evolved via this perfect fractal

growth evolution by chance? Had there been a key reversal today, it

would have been a completely perfect pattern in concert with a

possible major reversal point for the Wilshire. 27 October 2005 will

be a most interesting trading in this context.

An intriguing possibility exists that the second fractal for the

equity composite Wilshire has already been completed. A 19/48/ 10 of

45-48 fractal decay pattern is now evident to this myopic fractalist.

In this fractal solution 35-38 more trading days exist to a primary

bottom with nonlinear devalutions of potentially massive proportions

contained within those 35-38 days.

A slope taken from the first day of the second fractal to the last day

of the second contains all of the lowest values for the intermediate

days - except 2-3 days. And even those days touch the underlying

slope line at the bottom most areas of their daily valuations.

By repetitive inductive and retrospective fractal analysis,

macroeconomics appears to be a recurrent cyclical process of natural

growth, saturation , and inevitable natural decay flowing in

recurrent, nearly quantum, maximally efficient fractals of incremental

time proportionalities. Growth and saturation represents the bulk of

the cycle. The terminal devolutions are a very small part of the total

economic life cycles; they are nonlinear; and the declines are

proportional to the antecedent excesses in growth, debt, and

overvaluations at the summit saturation levels.

Gary Lammert http://www.economicfractalist.com/

Posted by: gary lammert | Thursday, October 27, 2005 at 09:01 AM

The holy grail. the construct that will serve ecometrics and provide a utilatarian yardstick for macroeconomics is:

economic fractal analysis; it is real and it is testible and operative:

The fractal decay puzzle which retrospectively will epitomize

maximum mathematical efficiency in the three fractal primary decay process must be solved prior to its evolution in order to validate fractal analysis as a fundamental 'science' of macroeconomics.

At its smallest time units, equity valuation growth to trading

saturation points and decay to selling saturation points makes

intuitive sense. At the longest time units, growth to the consumer debt saturation point based both on collective limited wages and overvaluation and overproduction of assets which in turn are based both on low interest rates and creative easy lending practices is followed by a contraction and decay process of asset values that unmasks the surpluses in the system and results in job contraction and lower wages based on those ongoing surpluses and contracting prices.

This longer perspective likewise makes intuitive sense.

Equity valuation fractals quantify, in the most efficient manner, this qualitative intuitively reasonable process. The markets are not - theyare not random walks. The markets are defined by dollars available for investment, current total value of investments and assets, ongoing wages, outstanding debt obligations, and inflationary pressures of day

to day living fueled by low interest rates, imprudent lending

practices, and inappropriate money creation in relation to traditional debt to wage ratios. The numbers 1-10 and fractions and multiples thereof define the ongoing valuations, debt, and money supply that is the market. That the market 'marches' and otherwise evolves to a beat of efficient fractal patterns that represents the total integration of these numbers into repetitive maximum efficient patterns is wholly reasonable.

The probability that the retrospectively identified recurrent fractals patterns occur by chance alone approaches zero. The challenge remains to decipher the non complex puzzle and to delineate the decay process prior its real time evolution. If it can be done, it will be the first time that quantification of the valuation decay process has been accomplished before the event. Ludwig Von Mises and Joe Kennedy by

different methodologies in 1928 and 1929 intuitively sensed in a qualitative way the lurking decay process and collapse and acted accordingly.

Just as in 1929 and 1720, consumer saturation and/or asset

overvaluation relative to ongoing wages and debt load will be the precipitating cause of the mechanistic deterministic non stochastic fractal decay process. By lowering the fed fund rates to extreme low levels and not controlling lending practices, the Federal Reserve and government caused excess borrowing, excess valuation, and excess production - with expected excess pain in the subsequent natural and mechanistic devolution process. By its ongoing raising of fed fund rates the Fed has placed boron rods into the uncontrolled reactor and

decreased the highs of overvaluation with an expected lessening of pain in the devolution. Hence the Euro- Nikkei markets with relatively less tightening by their central banks have had higher recent terminal valuations. Regardless of the degree of overvaluation, devolution will inevitably occur and will inevitably occur on time according to the most efficient fractal decay progression Just as the Federal governmen and the Federal Reserve with its large check book and rapid lowering of interest rates, respectively, could not change the couse of the devolution in 1929-1932 and 2000-2003, so will they be powerless to change the course of the present devolution.

The perfect often stated fractal growth pattern is x/2.5x/2x followed by an idealized decay of 1.5x. This last 1.5x can be subdivided into an idealized decay fractal pattern of y/2.5y/2.5y.

TNX, the ten year note, had an exhaustion gap on Wednesday 26 October, day 38 of a 19/48/38 of 38 perfect growth sequence. What is possibility that this perfect x/2.5x/2x growth fractal with an exhaustion gap on the last day of the third growth fractal evolved via this perfect fractal growth evolution by chance? Had there been a key reversal today, it would have been a completely perfect pattern in concert with a

possible major reversal point for the Wilshire. 27 October 2005 will be a most interesting trading in this context.

An intriguing possibility exists that the second fractal for the

equity composite Wilshire has already been completed. A 19/48/ 10 of 45-48 fractal decay pattern is now evident to this myopic fractalist. In this fractal solution 35-38 more trading days exist to a primary bottom with nonlinear devalutions of potentially massive proportions contained within those 35-38 days.

A slope taken from the first day of the second fractal to the last day of the second contains all of the lowest values for the intermediate days - except 2-3 days. And even those days touch the underlying slope line at the bottom most areas of their daily valuations.

By repetitive inductive and retrospective fractal analysis,

macroeconomics appears to be a recurrent cyclical process of natural growth, saturation , and inevitable natural decay flowing in recurrent, nearly quantum, maximally efficient fractals of incremental time proportionalities. Growth and saturation represents the bulk of the cycle. The terminal devolutions are a very small part of the total economic life cycles; they are nonlinear; and the declines are proportional to the antecedent excesses in growth, debt, and overvaluations at the summit saturation levels.

Gary Lammert

Posted by: gary lammert | Thursday, October 27, 2005 at 09:54 AM

Gary,

I notice that you run your X, 2.5X, 2X FROM BOTTOM TO BOTTOM. cAN THE SAME THING BE DONE FROM TOP TO TOP?

Posted by: Dennis Belanger | Saturday, November 12, 2005 at 05:07 PM

Dad, you are too smart for your own good. ~Lo

Posted by: Laura Lammert | Sunday, February 26, 2006 at 06:06 AM