The Goal of this Project
Given the fact that the staggered implementation of EDGAR can be considered as an exogenous shock to information dissemination technologies, we would therefore apply difference-in-difference (DID) analysis framework. In general DID framework, we will split the whole samples into two groups, one of which receives treatment and the other serves as control group, and therefore we can compare these two groups and check if the treatment really makes differences. This DID framework is known to be a nice econometric model to overcome endogeniety problem that is a common for most of empirical finance researches.
However, an important premise for DID needs to be fulfilled; that is: the only difference between treatment and control groups should be the treatment itself. To put it differently, if there is another confounding factor that also affects treatment group, then the difference between two groups could be driven by this confounding factor instead of the treatment we are focus on.
Therefore, the goal of this project is to check if there are any confounding factors and make sure if our data are random samples, from the perspective of data visaulization. We will investigate in two potential scenarios.
- Whether all the funds are assigned into 6 gorups by SEC randomly?
- Whether the treatment and control groups differ in other fund characterisitics, such as fund size, fund flow, or fund ages?
We will answer the above questions in the later two sections.