跳至正文
返回文库全部文档

交易价值与交易量、市场波动及预测限制之间的关系

文章 arXiv papers · 作者: Victor Olkhov

总结

本文提出,市场交易中的随机性会促成价格和收益的随机行为。文章考察一个平均区间内的交易价值和交易量,将其视为随机变量,并把价格和收益波动率与这些变量的波动率及相关性联系起来。文章还将这一视角延伸至宏观经济变量的准确性限制,并以宏观经济投资准确性为例。

文章认为,常见宏观经济模型关注交易价值或交易量总和等一阶量,而预测波动率需要能够表示涉及二阶变量之依赖关系的理论。文中主张,缺少这类理论会限制波动率预测的经济学依据,并使预测准确度至多受高斯分布约束。这是摘录中的概念性主张:它没有提供实证分析、估计程序,也没有比较不同模型的证据。因此,所提出的关系仍是进一步收集数据和开展计量经济学研究的动因,而非已经验证的预测方法。

核心观点

  • 本文将交易价值和交易量中的随机性视为价格与收益波动的来源。
  • 文章将市场波动与平均区间内交易活动的变化和相关性联系起来。
  • 文章认为,宏观经济模型需要考虑二阶变量之间的依赖关系,才能预测波动率。
  • 摘录呼吁采用计量经济学方法、收集数据并发展理论,但未提供实证检验或实用预测流程。
  • 文章主张,在缺少所述二阶宏观经济理论的情况下,预测准确度受高斯分布限制。

标签

全文
# Volatility Depends on Market Trades and Macro Theory


# Volatility Depends on Market Trades and Macro Theory









We consider the randomness of market trade as the origin of price and return stochasticity. We look at time series of trade values and volumes as random variables during the averaging interval Δ and describe the dependences of market-based volatilities of price and return on the volatilities and correlations of market trade values and volumes. We describe the market-based origin of the lower boundaries of the accuracy of macroeconomic variables and consider, as an example, the accuracy of macroeconomic investments. We highlight that current macroeconomic models describe relations between the 1st order variables determined by sums of trade values or volumes. To predict market-based volatilities of price, return, and volatilities of macroeconomic variables, one should develop econometric methodologies, collect data, and elaborate macroeconomic theories of the 2nd order that model the mutual dependence of the 1st and 2nd order economic variables. The absence of macroeconomic theories of the 2nd order means no economic basis for predictions of market-based volatilities of price and return, as well as volatilities of any macroeconomic variables. In turn, that limits the accuracy of forecasting probabilities of price, return, and the accuracy of macroeconomic variables in the best case by Gaussian distributions.

在遵守原作品许可的前提下,附作者信息全文展示。 许可协议: abstract CC0

此摘要由 Stratmill 研究智能体根据原文撰写,并非原文副本。