Towards a Descriptive Model of Agent Strategy Search--论文代写范文精选

与复杂的系统观察和描述来说,只有详细的行为机制很好理解，似乎太多的经济行为是复杂的。这不会很奇怪,因为它产生的结果之间复杂的相互作用，导致其行为的多样性。下面的essay代写范文进行详述。

Abstract
It is argued that due to the complexity of most economic phenomena, the chances of deriving correct models from a priori principles are small. Instead are more descriptive approach to modelling should be pursued. Agent-based modelling is characterised as a step in this direction. However many agent-based models use off-the-shelf algorithms from computer science without regard to their descriptive accuracy. This paper attempts an agent model that describes the behaviour of subjects reported by Joep Sonnemans as accurately as possible. It takes a structure that is compatible with current thinking cognitive science and explores the nature of the agent processes that then match the behaviour of the subjects. This suggests further modelling improvements and experiments.

Introduction 
Complex systems are precisely those for which it is extremely difficult to deduce its behaviour from first principles1. For example, it is extremely unlikely that one would be able to predict the behaviour of a meercat purely from a priori principles, rather one would have to spend a lot of time observing and describing its actions to get a hold on the intricate contingencies of its actual behaviour. With complex systems observation and description must come first and only much later when the detailed behavioural mechanisms are well understood is it sometimes possible to encapsulate some of these in a predictive model. It seems likely that much economic behaviour is complex in this way. 

This would not be very surprising since it arises as the consequence of the intricate interactions between members of a species that is characterised by the variety and contingency of its behaviour. But if we are to give up the chimera of numerical predictive models built from a priori principles, doesn’t that mean we have to give up all formal models and rigour? I would say that we do not. What it does mean, however, is that we have to use formal and computational models that are able to reflect the detailed behaviour as it is observed. We need to constrain our models as much as possible using observations of the relevant phenomena, both in terms of the causal processes as well as the outcomes. Pinning down our models using only the verification of predictive outcomes and an insistence on formal simplicity will not be enough. 

We will need to capture the workings of the processes stage by stage as they are observed and reproduce the known outcomes. In order to perform this feat we will need formal systems that are up to the task of expressing the qualitative cognitive processes that economic processes are rooted in. These more expressive systems come with a price, they are not simple and they seem to allow for multiple representations of the same outcomes. However there is no need for them to be any less formal or rigorous than a set of differential equations. In this paper I will exhibit an attempt to construct a more descriptive model of the search for an appropriate strategy by the subjects in a particular experiment. It is, of course, impossible to lose all assumptions in the construction of any model, but the point is to move towards using fewer and less drastic a priori assumptions and use more qualitative and quantitative constraints derived from the processes under study. The purpose of this model is to provide an unambiguous framework for the exploration of the possible processes within these constraints so as to inform the direction of further observation and modelling. 

This is not merely a static description, for I am not concerned with static phenomena, but a dynamic description of a particular set of observations using the techniques of computational and cognitive modelling. The extent to which this model is generalisable to other phenomena will only become apparent when it is compared with other descriptive models, just as the general characteristics and markings of a species of animal may only become clear when several descriptions of the animals are compared. To many readers my position will seem too pessimistic. They may be still hoping for some brilliant ‘short-cut’ to a predictive model, that will allow them to miss out the laborious business of observing and describing the underlying processes. However, I would point out that the science of biology has become enormously successful using the methods I am suggesting and, once we have accepted the amount of field work that our subject matter entails, equal success might be achieved in economics.

Computational, agent-based models 
The move to agent-based models in economics can be seen as part of a transition to a more descriptive style of modelling. An agent-based model must, by its very nature, model a real actor with a computational agent (in some way), so there should be a one-one correspondence between actors and agents. It is not necessary to assume that the law of large numbers will iron out the messy details. The model can allow the global properties to emerge (or not) without having to assume these details away. Real economic actors are (almost always) encapsulated, i.e. they will have an inside where the decision making is done which is largely hidden from view, and a series of ways in which they interact with the outside environment which are observable. The agents that are used to model these actors are encapsulated in a directly analogous way1.

However, many agent-based modellers do not see the need for any greater descriptive accuracy than this. Thus when inspecting the learning, inference and decision making processes that an agent uses in such a model, one often finds something as unrealistic as a simulated annealing algorithm or standard genetic algorithm. These are algorithms that have been taken from the field of computer science, regardless of their descriptive appropriateness for the actual economic actors being modelled. Now it is possible that in some circumstances such algorithms will give acceptable results for the purposes of some models, but at the moment we can only guess whether this is the case. It is not only that we do not know the exact conditions of application of each algorithm, we do not know of even a single real circumstance where we could completely rely on any of these ‘off-the-shelf’ algorithms to give a reasonable fit. 

To be clear, I am not criticising looking to computer science for ideas, structures and frameworks that might be used in modelling. Being a bounteous source of possible types of process is one of the field’s great contributions to knowledge. What I am criticising is the use of such algorithms without either any justification of their appropriateness or modification to make them appropriate. Thus many agent-based models fail to escape the problems of more traditional models. They attempt to use some ensemble of interacting agents to reproduce some global outcome without knowing if the behaviour of the individual agents is at all realistic. The wish for the ‘magic’ short-cut is still there. Clearly what is needed is some way of modelling the behaviour of economic actors by computational agents in a credible way. As noted above, real economic actors are probably complex in the sense that it is unlikely that we will be able to deduce their actions from a priori principles. What we can do is to constrain our models as much as possible from what we know. There are several sources of such knowledge.

The target behaviour 
The behaviour I am aiming to capture in my descriptive model is that specified by the results of ‘experiment 1’ described by Sonnemans in [12]. This is an experiment where subjects have to repeatedly sell a notional item in a bidding process. They receive a sequence of offers for this item. Each offer costs them 2 cents and each offer is drawn randomly from the interval [1,100] – they are paid the amount of the offer that they accept minus the costs incurred (2 cents per offer accepted) minus a fixed fee of 50 cents per game. Thus they have opposing incentives: to wait for a better offer and to avoid excessive cost. The ‘optimal’ strategy is to have a reserve price, that is wait for a offer of 81 or over and then accept it. The experiment is divided into parts: 0, 1, 2, 3, and 4, each of 4, 15, 15, 15 and 20 games respectively. 

Part 0 is the practice stage where the agent learns but statistics are not kept and no earnings gained. In parts 1, 2 and 3 the agents learn and earn as they do so. In part 1 the game starts in earnest, so that the subjects earn real money dependent upon their performance. At any stage in the game the subjects have the option of finding out any combination of the following information about the game so far: the number of bids; the last bid; the highest bid so far; the cost of bids if they stopped; and the earnings if they stopped. Part 2 is the same as part 1, except that at any stage the subjects can only access one of the above pieces of information (this does not stop them remembering or working out this information in their head, of course). In part 3 the first 0, 1, 2, 3 or 4 offers (determined randomly) were automatically accepted, the subject deciding when to stop after that (these offers still had to be paid for). 

In the last part page 5 (part 4) a constant strategy (the best learnt by the end of part 3) is kept for each game, so no learning occurs but statistics are kept and payments made. What makes this particular experiment appropriate for this purpose is that Sonnemans extracts the strategies that the subjects end up with in a form that is computationally modellable. In the experiment there were several parts. In the initial parts the subjects were able to try out their strategies. Before the last part they had to specify the strategy that would determine offer acceptance for the final 20 games. In all but two cases these could be formulated in terms of five predicates and two Boolean operators. This is not surprising as Sonnemans had done a pilot study to determine the operators that most people would use. These were: H≥x (the highest offer so far is not less than x); L≥x (the last offer was not less than x); N≥x (there have been x offers or greater); E≥x (earnings are not less than x); O≥x (there have been at least x offers in a row since the last highest offer or more); AND (boolean conjunction); and OR (boolean disjunction). Thus the strategy ‘Accept the highest offer if my earnings are at least 70 or there have been 10 offers’ could be expressed as ‘H≥70 OR N≥10’.

51Due网站原创范文除特殊说明外一切图文著作权归51Due所有；未经51Due官方授权谢绝任何用途转载或刊发于媒体。如发生侵犯著作权现象，51Due保留一切法律追诉权。
更多essay代写范文欢迎访问我们主页 www.51due.com 当然有essay代写需求可以和我们24小时在线客服 QQ:800020041 联系交流。-X