Price discovery and market microstructure noise

• Autor(a) principal: Fernandes, Marcelo
• Data de Publicação: 2019
• Outros Autores: Dias, Gustavo Fruet, Scherrer, Cristina Mabel
• Tipo de documento: Artigo de conferência
• Idioma: eng
• Título da fonte: Repositório Institucional do FGV (FGV Repositório Digital)
• Texto Completo: https://hdl.handle.net/10438/27734
• topic: High-frequency data; Component share; Information share; Market microstructure noise; Instrumental variables; Weak instruments; Realized measures; Economia; Teoria da informação em economia; Econometria; Análise de séries temporais

Using a continuous-time price discovery model, we show that the standard econometric framework typically yields inconsistent estimates of the price discovery measures in the presence of market microstructure noise due the usual errors-in-variable problem. We address this issue using instrumental variables. We devise two sets of valid instruments for alternative assumptions on the market microstructure noise, and then establish consistency and asymptotic normality of the corresponding price discovery measures. We illustrate our findings by investigating price discovery for Alcoa, showing that market leadership conclusions depend heavily on whether we account or not for market microstructure noise.