psychoceramics: [Re: Economics and Psychology--5:: Emotive Equation]

[Peter Hipwell <petehip @ cogsci.ed.ac.uk>]

This was in sci.cognitive, cross-posted to a bunch of other groups.

Is it kooky, or is it just typical psychoeconomobabble? Or both? 

"Fertilizer Positive Utility"?

> >
> >							13 June 1996
> >
> >
> >		   ECONOMICS AND PSYCHOLOGY--5:: Emotive Equation
> >
> >				(Follow-on Article)
> >
> >				      
> >			   A Correspondent's Comparison 
> >		of the Emotive Equation With Neoclassical Theory
> >				     - Plus -
> >			      Responses to Questions
> >
> >
> >	A USENET reader wanting to determine new results
> >permitted by the Emotive Equation has provided an analysis of the
> >theory in relation to standard (neoclassical) theory, and offered
> >several observations and questions. He has suggested, and I agree,
> >that other readers may have the same questions. With the
> >correspondent's concurrence I accordingly post his letter with my
> >responses. 
> >	Before going to the topic, a few introductory comments
> >appear beneficial.
> >
> >	The Emotive Equation represents the expectational
> >(intertemporal) plan of the individual, accounting for expected
> >uncertainty, emotive mapping of intertemporal utility into real-time
> >anticipatory pleasure, and the expected constraints on behavior along
> >each of the infinite number of worldlines of the plan. (See Note 1.)
> >As could be expected of a comprehensive and coherent theory of
> >behavior, the methodology has demonstrated an important new
> >explanatory capability. For example, the theory has been shown to
> >determine the net marginal return on investment over investment for
> >two prominent models of capital -- Solow's Costless Conversion
> >Model (1965) and Lange's Two Production Function Model (1935).
> >(See E&P-4j and E&P-4p in sci.psychology.theory or
> >alt.economics.austrian-school.) Because this theory properly
> >(explicitly) accounts for time in its canonical formulation, it can
> >realistically model dynamic behavior at all levels 
> >-- this in contrast to standard (neoclassical) theory wherein the
> >incorrect (i.e., direct) assignment of utility to consumables
> >undermines time as an essential parameter (e.g., the time constraint,
> >arguably indispensable to economics as a science, is precluded).
> >	
> >	In the usual format, the correspondent's text is indicated by a
> >symbol (>) at the left of each line. Parts of my answers are placed
> >within the correspondent's paragraphs, and these insertions are
> >identified by enclosing brackets.
> >
> >				----------------
> >
> >>I'm sending this to you directly as newsgroups have the tenor of a
> >>thesis defense whereas I just want to understand what you've posted.
> >>I know I'm imposing so I don't expect a direct response, but maybe
> >>this will be representative of questions that others have also. 
> >>There are two parts to the following. First I give my understanding
> >>of your idea by summarizing the conventional approach and
> >>contrasting yours to it. I then ask some questions re specific
> >>postings. I'm sure my presentation isn't an exact description of your
> >>position, but I hope it's close enough that you will be able to discern
> >>from my questions where I'm not following your argument
> >>correctly.
> >>
> >>Let me preface my remarks by noting that, as I use the term here,
> >>models that do not differ in their response to stimulus are not
> >>different models. That is, even though they may have been derived
> >>from different concepts, if they measure the same they are the same.
> >>For instance, I would categorize all filters the same (whether analog,
> >>digital, or hybrid) if they have identical impulse responses.
> >>
> >>I start with a premise which I believe is generally held, that a
> >>mathematical model of the consumer's decision-making process, i.e.
> >>a utility function and associated behavioral rule, are essential to
> >>understanding economics as a science. This underpinning goes by
> >>many names; choice-theoretic, methodological individualism, or
> >>micro-foundations, for instance. It is understood that an individual
> >>performs the critical operations of decision-making in a
> >>psychological domain -- his mind. The role of the model is to
> >>provide a mapping of this unobservable and personal process to a
> >>real, that is economic, domain in such a way that the interactions
> >>between the individual's decision making and the economic
> >>process he participates in, may be observed.
> >>
> >>For macroeconomic studies, the model is used to evaluate the effects
> >>of changes in the economic system, such as prices or wages, and
> >>taxes or natural resources (that is, changes to endogenous and
> >>exogenous parameters, respectively) by observing the response of
> >>the individual and taking it to be representative of all economic
> >>agents. The idea being that while the domain where a consumer's
> >>decisions are made can't be observed, his response to a stimulus in
> >>the economic domain can be and, for an economics study, that is all
> >>that is needed. (There is the additional assumption, not at issue here,
> >>that agents' effects may be linearly combined to represent the
> >>totality.)
> >>
> >>The neoclassical paradigm is constructed on the assumption that an
> >>individual is able to select between alternative bundles of products
> >>and services (economic goods) in the context of two types of
> >>constraints. One is purchasing power, for instance his income, and
> >>the other is information, typically current and past prices. His
> >>ordering of available alternatives is called his "preferences" and the
> >>utility function and behavioral rule compute the value of these
> >>preferences in real terms. The model does this by converting a
> >>preference equilibrium condition, i.e. "indifference", to
> >>equivalence of goods "at the margin". The overall economic process
> >>denominates this equivalence in money terms.
> >>
> >>As I understand your criticism of this paradigm, you believe that the
> >>neoclassical model mis-specifies the basic determinant of the
> >>consumer's decision process. It pictures the individual choosing over
> >>goods, whereas you believe he chooses over the amount of time he
> >>will devote to certain categories of activities -- productive labor,
> >>consumption, and leisure. You also hold that an essential element of
> >>every decision is a subjective weighting, by the individual, of a
> >>decision's future consequences where this subjective weighting is
> >>implemented as anticipatory pleasure/displeasure.
> >>
> >>The questions came up as I attempted to investigate [a colleague's]
> >>query regarding what new results obtain from your formulation. It
> >>would seem that by changing the choice variable from goods to
> >>activities, you have 'endogenized' time which potentially has
> >>interesting consequences. Accordingly, I am trying to write the
> >>agent's maximization problem in some way that I can look at
> >>dynamic programming, the Euler equation, etc. as appropriate.
> >>Attempting to state the agent's choice problem I ran into some things
> >>I couldn't get a handle on:
> >>
> >> Re Econ and Psych 3: How does the relation P = dU/dt imply
> >>behavior? Taking it as an identity, should I understand it as the
> >>agent is a pleasure-seeker; that he maximizes pleasure through those
> >>selections that exhibit the maximum utility time intensity? [No, the
> >>agent does not seek to maximize intertemporal pleasure. (See
> >>below.)] Alternately, do I integrate both sides and consider the agent
> >>a utility maximizer where he searches for the best pleasure-time
> >>[integration]? [Correct, except emotive mapping participates. See
> >>{a}.]
> >
> >{a}	The relation P=dU/dt is analogous to  F=ma  in physics. The
> >latter is a principle of physical reality on which, for instance, the
> >Navier-Stokes Equations of fluid mechanics are formulated
> >(accounting for the effects of mass density, pressure gradients, shear
> >forces, etc.). P=dU/dt is a principle of psychical reality on which the
> >Emotive Equation is formulated (accounting for expected uncertainty,
> >emotive mapping, and expected constraints). Regarding the questions,
> >it is dE=MPdt rather than dU=Pdt that is integrated along the length
> >of each worldline of the concerned candidate expectational
> >(intertemporal) plan, where M is the Emotive Mapping Function.
> >(dU=Pdt represents what the individual expects to experience at
> >intertemporal time t on the concerned worldline, while
> >dE=MdU=MPdt represents the individual's real-time differential
> >pleasure experienced in anticipation of dU=Pdt.) The integral
> >expression is written:
> >
> >				  z					
> >		E	=	{} [	M(t)	P(t)	] dt		
> >				 0					
> >
> >(Ignoring worldline probability and the multiplier terms. z represents
> >infinity.) Note that the integral of the emotively mapped differential
> >utility along each worldline is not maximized independently of the
> >individual's expectation along the remaining worldlines of the
> >candidate intertemporal plan. The reason is, of course, that worldline
> >coincidence -- particularly at and near the start of the intertemporal
> >interval -- is an essential property of the methodology. (In this regard,
> >as intertemporal time expectedly proceeds, the individual's expected
> >activities along each worldline become progressively more unique --
> >i.e., the degree of worldline coincidence diminishes).
> >
> >>If I take the agent's plan of future time allocations as his choice
> >>variable (as discussed in the last paragraph), what is invariant?
> >>[(Expectational) worldline occurrence probabilities, emotive
> >>mapping functions, P-O-N pleasure functions, and constraints, for
> >>the considered plan. See {b}].  May I assume that he has fixed
> >>criteria for selecting the path such that, confronted with the same
> >>subjective assessment of the probabilities of future states, he will
> >>make identical time allocations? [Yes. See {c}.] Also, is the
> >>selection conditional on his accumulated experience up to that
> >>point? [Yes. See {d}.] If so, does that experience enter objectively
> >>or subjectively? [Both objectively and subjectively.]
> >
> >{b}	In the solution process, all unknowns within the control of the
> >individual are variable -- future time allocations, differentiable
> >capital, differentiable products (factors and consumables), savings,
> >etc. Additionally, the seller in the market sets the prices of the
> >supplied goods and services in accordance with his overall
> >expectational plan, such prices including  loan interest (and deposit
> >interest). It is seen that the Emotive Equation, in its application, can
> >substantively represent economic life, this in contrast to the ill-
> >founded (Walrasian) neoclassical theory.
> >	It is reiterated here that while all individual-controlled
> >unknowns are variable, intertemporal utility arises only from the
> >integration  of  pleasure  over  time  --  i.e.,  utility  does  not  arise
> >from  the  integration   of   consumable   specific   utility
> >([PLEASURExTIME/GOOD]) over a quantity of the consumable
> >(utility is imputed to the good, and not directly assigned.) 
> >{c}	Having expectationally defined the worldlines and time
> >allocations of a candidate plan, when confronted with the same
> >(expectational) assessment of the worldline probabilities, P-O-N
> >pleasure functions, and emotive mapping functions, the individual
> >may be assumed to make identical time allocations.
> >{d}	Expectational plan formation, including the resulting
> >anticipatory pleasure, is profoundly dependent on accumulated
> >experience (e.g., education, advertising, peer relations, etc.). It is also
> >dependent, of course, on the present, real-time physical, mental, and
> >emotional condition of the individual. (It may be noted here that the
> >rigorous view must be that any departure, however small, from
> >expected experience -- i.e., any real-time surprise -- initiates a new
> >expectational plan. Of course, we need not be so precise in applied
> >theory.)
> >
> >>(2) Re Econ and Psych -- Emotive Equation: In equations [4] and
> >>[5], how are "worldlines" and "plans" distinguished? [See {e}.] Are
> >>the worldlines the possible states which the agent assesses but nature
> >>selects, while plans are the agent's choice variable? [Yes. See {f}.]
> >>Is [4] meant to represent the domain of the agent's choice? [See
> >>{g}.] Does he select that plan, k, which maximizes utility, U^I? [He
> >>selects that plan which provides the greatest E^I or E^i -- the latter
> >>for the socioeconomically interactive individual.] Is the plan k the
> >>agent's mechanism for maximizing utility? That is, given probable
> >>future state sequences (worldlines) w -- possible states which he has
> >>subjectively weighted for probability of occurrence -- does he
> >>construct a plan by choosing how he allocates his time in each day
> >>for each sequence? [Yes. See {h}.] 
> >
> >{e}	Candidate (competing) expectational plans are separate or
> >distinct, and each candidate plan has a corresponding (infinite) set of
> >coupled worldlines. 
> >{f}	In preparing an intertemporal plan the individual defines the
> >corresponding worldlines -- each comprised of a continuous sequence
> >of activities, with associated P-O-N pleasure functions, emotive
> >mapping functions, and constraints, referred to as a state -- and nature
> >specifies which worldline the individual will follow (see Note 2). The
> >individual's experience will follow the selected worldline until the
> >operative plan is negated by surprise, at which time a new plan is
> >initiated.
> > {g}	The agent determines the expected future into which he or she
> >will proceed. This is accomplished by first preparing one or more
> >candidate plans, each with a corresponding (infinite) set of worldlines
> >with selected activity durations that maximize total intertemporal
> >utility as presently evaluated -- i.e., plan anticipatory pleasure. (This
> >is done on the basis of the Emotive Equation, Eq. [5] -- not Eq. [4],
> >which is an intermediate step in deriving Eq. [5].) To briefly
> >elaborate, the individual chooses the candidate plan  that provides the
> >greatest pleasure in its anticipation. In the market economy, this plan,
> >as real-time proceeds, determines when and how long the individual
> >will work to acquire money to purchase goods, when the purchases
> >are made and the amounts thereof, and the timing and rates of
> >consumption/utilization of the goods. All expectationally recognized
> >goods, productive and consumptive, receive imputed anticipatory
> >pleasure (see Note 3). It is seen that product demand and product
> >prices are intertemporal plan dependent. 
> >{h}	It may be understood in applied theory that the worldlines
> >(except for activity durations) are first defined -- where the worldlines
> >include (expectational) occurrence probabilities, emotive mapping
> >functions, P-O-N pleasure functions, and constraint relations -- after
> >which the optimal activity durations along all worldlines are
> >(simultaneously) determined, along with the corresponding product
> >purchases and sales at expected prices, etc. Note that when the
> >purchase or sale of one or more integral (nondivisible) products is
> >considered, the individual will choose between alternative
> >expectational plans. (Note 3.)
> >
> >>Also, what is the equivalent of the capital-based production in the
> >>model of Econ and Psych--4p? That is, what is the mechanism
> >>forcing an inter-temporal relationship in the equations?
> >
> >	The mechanism is the individual's expectation that capital
> >used in the production of itself and food will be carried over into 
> >day 2 -- i.e., if there were no expected capital carryover (e.g., w=0), 
> >the expected day 1 and day 2 activity durations would be independently
> >determined. Note that in the case where w=0 the expected normalized
> >return on investment at the margin (NMRetI/I) would be zero. 
> >
> >>(3) Re Econ and Psych--4p: In the discussion following the
> >>investment formula you seem to suggest that assigning expectational
> >>probabilities (the value of "a") should be based on counting, i.e.
> >>classical probability theory. Does that mean Bayesian estimation is
> >>not valid? I ask this because it would seem the subjective a priori
> >>probability belief (the "prior") in a Bayesian formulation is a natural
> >>fit to your approach since it combines experience (frequency) with
> >>intuition (subjective expectation), plus learning.
> >
> >	It was not intended that any particular understanding of
> >expectational probability be adopted, and the Bayesian formulation
> >would be suitable.
> >
> >			--------------------------------
> >
> >	In concluding this post, the reader is referred to newsgroup
> >sci.psychology.theory or alt.economics.austrian-school for reposts of
> >selected articles in the Economics and Psychology series over the past
> >seven months. A list of the posted articles is provided in Note 4
> >below.
> >
> >Tom Chamberlain, PhD/ME
> >
> >
> >NOTES:
> >
> >(1)	The Emotive Equation, shown below, was introduced to the
> >scientific community (via USENET) for the first time in early May.
> >This equation, all parameters and terms of which are expectational,
> >converts a differential utility (Pdt) at an imaginary time on any given
> >worldline into a real-time differential anticipatory pleasure, the result
> >being integrated to infinity along the given worldline and summed
> >over all worldlines, accounting for emotive mapping M and
> >uncertainty f, to produce the pleasure E that the individual receives in
> >anticipation of the considered candidate expectational plan. The
> >individual is purposeful in his expectational thinking, and determines
> >his intertemporal activity durations such that E is maximized (i.e., E
> >is an extremum, with dE = 0). Of the candidate intertemporal plans
> >that may be considered, the one that provides the greatest anticipatory
> >pleasure is chosen. The equation may be written:
> >	
> >			EMOTIVE EQUATION
> >
> >	 i	    	  i	   z	  i	 i
> >	E  =	[]	{f	{}	[M (t)	P (t)]dt
> >	 k	w=1,z	  kw	  0	  kw	 kw
> >
> >
> >						 ic	  ic		
> >				+	[]   (	L	F     )	}	
> >					c=1,z	 kw	  kw		
> >
> >where [] and {} represent the summation and integral signs, z
> >represents infinity, and  
> >
> >				 ic					
> >				F	=	0,	[Constraints]	
> >				 kw					
> >
> >for  all  individuals  i,  candidate  plans  k,  worldlines  w,  and
> >constraints c. L is the Lagrange multiplier for the corresponding
> >constraint.
> >	The selected (operative) expectational plan is corroborated by
> >actual experience until a new plan is commenced -- the operative plan
> >being terminated by surprise (due to an unexpected event, new
> >information, or a creative thought or thought process). 
> >	A   derivation   and   discussion   of  the  Emotive  Equation 
> >is   provided   in   E&P-5   (temporarily   posted   in 
> >sci.psychology.theory and alt.economics.austrian-school).
> >
> >(2)	All expectational plan worldlines must be coincident at the
> >start of the intertemporal interval. It is with the imagined progress of
> >intertemporal time that expected uncertainty effects the divergence of
> >worldlines.
> >
> >(3)	Imputation of an entity's value is performed by assessing
> >anticipatory pleasure for two plans, one with the entity expectedly
> >present and the second with the entity expectedly absent (i.e., without
> >compensation). The difference between the total anticipatory
> >pleasures of the two plans is the measure of the value of the entity to
> >the individual. If the entity has an incremental character -- for
> >example, as the margin of an infinitely divisible good -- its imputed
> >anticipatory pleasure, and value, can be determined as a perturbation
> >of the corresponding expectational plan. However, if the entity is
> >integral -- a car, for example, or an eye -- one compares the
> >anticipatory pleasures of the alternative (possibly radical different)
> >expectational plans to arrive at the difference in anticipatory pleasure
> >and, equivalently, the entity's (imputed) value.
> >
> >(4)	Selected articles from the "Economics and Psychology" series:
> >	(Currently reposted in sci.psychology.theory and
> >	 	alt.economics.austrian-school.)
> >	("Item" denotes a prominent subject in the article.)
> >
> >E&P-3: Behavioral First Principle. (10-17-95)
> >E&P-4: Thinking About The Unthinkable. (11-13-95)
> >E&P-4h: Item =Gossenian Approach: An Intuitive Example.
> >			(12-15-95; Orig.)
> >E&P-4j:  Item= Costless Conversion Model of Capital. (1-21-96)
> >E&P-4m:Item= Modeling of Pure Leisure. (2-4-96)
> >E&P-4n: Item= Fertilizer Positive Utility. (2-11-96)
> >E&P-4p: Item= Two Production Function Model 
> >			of Capital. (2-28-96)
> >E&P-5: Emotive Equation (5-7-96)
> >E&P-5a: Addendum (References, Notes, and Appendix) (5-7-96)
> >
> >(5)	One of the challenges in the mathematical modelling of
> >economic life is defining a nomenclature that is not so
> >burdensome as to be self-defeating. An approach in this regard is to
> >use "nested" superscripts and subscripts, e.g.,
> >
> >				   A-P		i			
> >			     <	  Dt	      >			
> >						[d-1]kw		
> >	
> >In this expression, the parameter Dt^A-P (where D signifies the
> >Greek symbol "delta") represents the {A}ppliance (i.e., capital)
> >{P}roduction activity duration of individual i in day d-1 along
> >worldline w of expectational plan k. The carrots serve to separate
> >superscript/subscript ensembles thereby aiding comprehension. Note
> >that an arbitrary number of terms for the same dkw of i may be
> >collected within opposing carrots.
> >
> >(6)	This article has been submitted to sci.psychology.research,
> >sci.psychology.theory, sci.cognitive, sci.econ, and
> >alt.economics.austrian-school.
> >
> >
> >REFERENCES:
> >
> >Gossen, H. H.  (1854) 1983  _ The Laws of Human Relations and
> >	The Rules of Human Action Derived Therefrom_.
> >	Translated by Rudolph C. Blitz with an introductory essay by
> >	Nicholas Georgescu-Roegen. Cambridge: MIT Press.
> >
> >Lange, O.  1936  "The Place of Interest in the Theory of
> >	Production". Review of Economic Studies. 3:159-192.
> >
> >Solow, R. M.  1965  _Capital Theory and the Rate of Return_. 
> >	Chicago: Rand McNally & Company.
> >
> >
> 
> 
> -- 
> --