What is it about?

It already is well accepted that 'consistency' is not the exact same thing as 'rationality'. Whereas, however, consistency is admitted to be a weaker constraint than rationality, hitherto there remained the anticipation, to wit, efforts for the maintenance of rationality are aided and abetted by effort that are applied to the maintenance of consistency. So then, if you were risk averse, the anticipation was, you were more likely to remain rational if you maintained risk aversion, that is, if you remained 'consistent'. Ditto if you are adventuresome, namely a socioeconomic agent who is adventuresome is more likely to maintain rationality if the agent remains adventuresome. The first chink in the armor of the anticipation, to wit, the effort for consistency aids and abets the maintenance of rationality appeared in Kreps and Ramey (1987). In stated respect, whereas Kreps and Wilson (1982) had assumed that the maintenance of rationality on a period-by-period basis is aided and abetted by effort for the maintenance of consistency, a rigorous examination of the hypothesis in Kreps and Ramey (1987) did not support the anticipation, that is, could not associate effort for the maintenance of consistency with the outcome, to wit, rationality also is maintained. Subsequent to that effort, there was arrival at a long silence on the subject. My study, newly published in the Elsevier Journal, 'Social Sciences & Humanities Open' breaks the long silence, shows any and all efforts that are geared towards the maintenance of consistency directly are inimical to the maintenance of rationality. Quite the contrary, the formal theoretical (mathematical) model arrives, explicitly at the finding that it is the openness (predisposition), in the current period, to the feasibility, to wit, the action that maintains rationality in the very next period is, in relation to the action that was undertaken in the current period, 'seemingly random' that maintains rationality. For illustration, suppose a socioeconomic agent is, in the current period characterized by risk aversion, that is, prefers certainty to taking on of a risk. Applying study findings, the agent allows for the maintenance of rationality 'if and only if', in the current period the agent is predisposed to becoming less risk averse in the very next period, as such is willing to take on a risk in the very next period. Subsequently, the agent tatonnes between 'more risk averse' and 'less risk averse' ad infinitum. For additional concreteness, suppose a socioeconomic agent is, in the current period characterized by adventuresomeness. The formal theory establishes that the agent allows for the maintenance of rationality 'if and only if', in the current period the agent is predisposed to becoming 'less adventuresome', equivalently is predisposed to becoming 'more prudent' in the very next period. Subsequently the agent tatonnes between 'more adventuresome' and 'less adventuresome' ad infinitum.

Featured Image

Why is it important?

An agent who attempts to maintain consistency does not search for new knowledge that, feasibly facilitates improvements to the decision that was undertaken in either of the current period or in historical periods. In a future period, however, an agent who, on the other hand, is willing to tatonne away from current or historical actions engages, necessarily with a juxtaposition of the different actions and objectives that, feasibly advance the agent's utility. Whereas then the first agent 'assumes' that the current action will remain the best action in future, as such necessarily maintains the same objective, the second agent engages, every period, with a search for each of the action and objective that best advances the agent's utility. As enumerated, the findings in my newly published study provide implementational insights into the findings in Obrimah (2022), which shows that rationality has parameterization as "non-satiation with respect to efforts that, feasibly unravel some unawareness that facilitates improvements to rational decision making." Whereas it might seem to some that study findings are 'much ado about nothing', several real life illustrations reveal that socioeconomic agents tend to gravitate towards consistency. First, if socioeconomic agents do not gravitate towards consistency, constitutions would be deemed 'fluid documents' that are amended as necessary. Very few countries treat their constitutions, as such. Second, if socioeconomic agents do not gravitate towards consistency, the U.S. Constitution would, in the event it were proposed by the polity, allow for a referendum as to whether the polity would allow a President who already has served two terms compete for a third. Third, if socioeconomic agents do not gravitate towards consistency, the facts of the current conflict between Russia and Ukraine would matter, as opposed to the knee jerk reaction, to wit, "it is wrong for one country to invade another." In stated respect, with former Chancellor Angela Merkel stating explicitly that the decision, on the part of NATO, to aggravate Russia was premeditated, and that she had objected to the action whilst she remained the German Chancellor, the notion that 'Russia is wrong' and 'Ukraine and NATO are right' is a wrong-headed argument. Add to the stated evidence, the decision, on the part of NATO, to violate the preexisting agreement, entered into at the end of the first Cold War, to wit, relative to each of Russia or NATO, Ukraine would remain a neutral country, and the notion that 'Russia is wrong' and 'Ukraine and NATO are right' downright is false. For concreteness, absent any real discussion of the merits of the war, which is evident in the fact that the United Nations is not making any effort to mediate between the two parties, the United Nations merely voted, 'to invade another sovereign country is wrong' - as such a vote for consistency, as opposed to the merits of the specific case. Personally, I yearn for the days of either of Kofi Annan or Boutros Boutros-Ghali, two leaders of the United Nations who typically were willing to take on the 'Big Countries'.

Perspectives

My study arrives at two important implications. First, whereas some believe that empirical tests are tests of the data that are analyzed, absent engagement with any econometrics, my publication demonstrates that empirical tests are tests of the models that are imposed on data, not tests of what resides in the data. In stated respect, all empirical results are the outcome of compositions of the data that are facilitated by the empirical model. Second, the enumerated inference results in a rational explanation as to why machine learning models which perform great in within-sample tests tend not to perform great in out-of-sample tests.

Dr Oghenovo A Obrimah
Fisk University

Read the Original

This page is a summary of: Should rational socioeconomic agents attempt to exert ‘control’ over the evolution of their preference or behavioral parameters?, Social Sciences & Humanities Open, January 2023, Elsevier,
DOI: 10.1016/j.ssaho.2023.100606.
You can read the full text:

Read

Resources

Contributors

The following have contributed to this page