Home > Uncategorized > Regression analysis — a constructive critique

## Regression analysis — a constructive critique

from Lars Syll

As a descriptive exercise, all is well. One can compare the average salary of men and women, holding constant potential confounders. The result is a summary of how salaries differ on the average by gender, conditional on the values of one or more covariates. Why the salaries may on the average differ is not represented explicitly in the regression model …

Moving to causal inference is an enormous step that needs to be thoroughly considered. To begin, one must ponder … whether the causal variable of interest can be usefully conceptualized as an intervention within a response schedule framework [a formal structure in which to consider what the value of the response y would be if an input x were set to some vaue]. Once again consider gender. Imagine a particular faculty member. Now imagine intervening so that the faculty member’s gender could be set to ‘male.’ One would do this while altering nothing else about this person …

Clearly, the fit between the requisite response schedule and the academic world in which salaries are determined fails for at least two reasons: The idea of setting gender to male or female is an enormous stretch, and even, if gender could be manipulated, it is hard to accept that only gender would be changed. In short, the causal story is in deep trouble even before the matter of holding constant surfaces …

This is not to imply that it never makes sense to apply regression-based adjustments in causal modeling. The critical issue is that the real world must cooperate by providing interventions that could be delivered separately …

As a technical move, it is easy to apply regression-based adjustmens to confounders. Whether it is sensible to do so is an entirely different matter …

The most demanding material [is] the examination of what it means to ‘hold constant’ … The problem [is] the potential incongruence between the mechanics of regression-based adjustments and the natural or social world under study.

1. October 3, 2018 at 3:44 pm

Until a clear distinction is made between concrete curve fitting of specific data and abstract theory, confusion will abound. Abstraction has to be based on first principles analysis. It would be blind chance to arrive at a correct equation form from most economic data. Time is not handled in an appropriate manner and is necessary this to be dealt with.

2. October 3, 2018 at 7:27 pm

Time as a factor can only be controlled in a system where that system terminally ends…like the point of retail sale in economics, and if at that point you implement a policy that beneficially resolves the two most chronic and relevant problems for every legitimate agent in the economy….that is the character and definition of a paradigm change.

• October 4, 2018 at 8:03 am

The passage of time implied by the second law of thermodynamics is uncontrollable.

Elapsed time expended in a project is under human control.

• October 4, 2018 at 9:21 am

The temporal universe under the laws of thermodynamics indeed is a continuously random but, if understood from a cosmic perspective, graciously flowing reality.

However, particular systems have their own particular and peculiar starting, changing and stopping points, and final retail sale is the terminal summing and stopping point for total costs and prices for every item or service for the legitimate economic/actually productive process and finally and consequently the terminal expression point for all forms of inflation.

The terminal stopping/summing point for costs and prices in economics is also an ultimate power point, And therein is a tremendously important but almost entirely missed significance, because at that point production becomes consumption, all agents come to a final agreement and no agent can be harmed or has legitimate recourse outside of fraud thereafter. Thus a digital 50% discount/rebate policy at that point has the power to change scarcity of individual income into a 100% increase/abundance/paradigm changing resolution to modern economy’s two most chronic problems ( individual income scarcity and further erosion of same by inflation) and yet does not disturb the sanctity so to speak of freely determined price because the rebate back to the enterprise giving the discount is made whole by it.

When economists and economic pundits can come up with a more sweepingly beneficial set of policies than the two I’ve posted here many times I’ll get on their bandwagon, and until they do they need to get off their “hobbie horse” theories, piecemeal reforms and their attachments to “long dead economic scribblers”.