Modeling the Emergence of Outliers in Entrepreneurship

这个项目将建立一个理论框架，以解决企业家最初的资源天赋、对新颖性的期望、与市场的接触以及一般市场环境如何相互作用，以促进极其成功和有影响力的新企业的出现。像这样的风险投资是异常值，因为它们很罕见，在多个层面上具有不成比例的影响。此外，这些离群点企业在投入和结果上都与大多数人所描述的典型企业截然不同。该项目的框架将增强关于组织系统如何在不断变化的竞争环境中随着时间的推移增长、改进绩效和实现异常结果的知识。因为这个框架建立在来自多个复杂社会系统的数据之上，所以它可以推广并能够改进个人、团队、公司、行业和国家的结果。该项目的主要研究重点是确定企业家精神中异常结果的主要驱动因素。在投入和结果呈正态钟形分布的假设下，很难建立关于异常值出现的理论。在正态分布假设下，异常值通常会从分析中清除--要么被删除，要么经过数学变换--以减少它们的影响。这个项目超越了正态分布假设，使用了复杂性科学的理论假设和方法。这种理论方法有两个优点。首先，从复杂性的角度来看，假设在相对不受约束的环境中，结果很可能会按照幂定律分布。这种分布是至关重要的，因为传统的理论构建和测试方法假设极端事件的概率实际上为零。幂定律假设可以更好地反映新兴企业研究中发现的经验分布。第二，复杂性观点认为，系统的结果主要是相互关联的投入组合的结果，包括1)代理人对资源的初始天赋，2)代理人基于对未来结果的预期与其他代理人接触的规则，以及3)环境资源的数量和质量。从这个角度来看，一个系统从不存在到存在是一个非线性的、以阈值为基础的过程。这个项目使用非参数最大似然Bootstrap估计的幂定律适用于新生的创业企业的代表性样本，以得出一个适用于所有组织形式的投入的元构造框架：捐赠、期望、参与和环境。然后，为了测试这个可推广的4E框架，这些输入被用来构建基于代理的仿真模型，其结果将通过来自增长最快的私人持股公司的经验数据进行验证。

This project will build a theoretical framework to address how an entrepreneur's initial resource endowments, expectations for novelty, engagement with the market, and the general market environments interact to facilitate the emergence of extremely successful and influential new ventures. Ventures like this are outliers because they are rare and have disproportionate impact at multiple levels. Moreover, these outlier ventures are very different in both their inputs and outcomes than what most would characterize as typical. The project's framework will enhance knowledge of how organizational systems grow, improve performance, and achieve outlier outcomes over time in changing competitive environments. Because this framework is built on data from multiple complex social systems, it is generalizable for and capable of improving the outcomes of individuals, teams, firms, industries, and nations. The overarching research focus of this project is to identify the primary drivers of outlier outcomes in entrepreneurship. Building a theory about the emergence of outliers is difficult under the assumption of normal, bell-shaped distributions of inputs and outcomes. Under the normal distribution assumption, outliers are usually cleansed from the analysis-either deleted or mathematically transformed-to reduce their influence. This project moves beyond the normal distribution assumption and uses theoretical assumptions and methods from complexity science. This theoretical approach has two advantages. First, a complexity perspective assumes that in relatively unconstrained environments, outcomes will likely be distributed according to a power law. This distribution is vital because traditional theory building and testing methods assume that the probability of an extreme event is effectively zero. A power law assumption can more adequately reflect the empirical distributions found in studies of emerging ventures. Second, a complexity perspective suggests that a system's outcomes are primarily the result of an interconnected combinations of inputs, including 1) an agent's initial endowment of resources, 2) the agent's rules for engagement with other agents based on expectations for future outcomes, and 3) the quantity and quality of environmental resources. Within this perspective, the emergence of a system from non-existence to existence occurs as a nonlinear, threshold-based process. This project uses non-parametric maximum-likelihood bootstrap estimates of power law fit on representative samples of nascent entrepreneurial ventures to derive a meta-construct framework of inputs that apply to all organizational forms: endowments, expectations, engagement, and environments. Then, to test this generalizable 4E framework, these inputs are used to build an agent-based simulation model, the outcomes for which will be validated with empirical data from the fastest growing privately held firms.