Learning in monopolies with delayed feedback on price expectations
Learning in monopolies with delayed feedback on price expectations
复制标题
通过对价格预期的延迟反馈来学习垄断
DOI：
10.1016/j.cnsns.2015.04.011
复制
发表时间：
2015
影响因子：
3.9
通讯作者：
Ferenc Szidarovszky
中科院分区：
文献类型：
--
作者：
Akio Matsumoto;Ferenc Szidarovszky
我们称价格函数与垂直轴的截距为最大价格，将价格函数的斜率称为边际价格。本文假设垄断企业具有关于边际价格和自身成本函数的完全信息，但对最高价格不确定。然而，通过与市场的反复互动，获得的价格观察为最高价格的自适应学习过程提供了基础。还假设价格观测具有固定的延迟，因此学习过程可以用时滞微分方程来描述。在一个或两个时滞的情况下，考察了所得到的动态过程的渐近行为，得到了稳定性条件。在两个延迟学习过程中得到了三个主要结果。首先，稳定单时滞模型中不稳定的平衡点是可能的。其次，涉及混沌的复杂动力学可能会出现，这在单延迟模型中是不可能的。第三，稳定和不稳定的交替(即稳定开关)反复发生。
We call the intercept of the price function with the vertical axis themaximum priceand the slope of the price function themarginal price. In this paper it is assumed that a monopolistic firm has full information about the marginal price and its own cost function but is uncertain on the maximum price. However, by repeated interaction with the market, the obtained price observations give a basis for an adaptive learning process of the maximum price. It is also assumed that the price observations have fixed delays, so the learning process can be described by a delayed differential equation. In the cases of one or two delays, the asymptotic behavior of the resulting dynamic process is examined, stability conditions are derived. Three main results are demonstrated in the two delay learning processes. First, it is possible to stabilize the equilibrium which is unstable in the one delay model. Second, complex dynamics involving chaos, which is impossible in the one delay model, can emerge. Third, alternations of stability and instability (i.e., stability switches) occur repeatedly.