Para receber por e-mail novas publicações de Seminário de Métodos Matemáticos em Finanças, Clique Aqui!

Seminários do IMPA

Métodos Matemáticos em Finanças

Forecasting Large Realized Covariance Matrices: The Benefits of Factor Models and Shrinkage
Marcelo Cunha Medeiros

Terça-feira, 5 de junho de 2018, 17:30
Sala 232

We propose a model to forecast very large realized covariance matrices of returns, applying it to the constituents of the S\&P 500  on a   daily  basis. To address the curse of dimensionality, we decompose the return covariance matrix using standard firm-level factors (e.g., size, valueand profitability) and use sectoral restrictions in the residual covariance matrix. This restricted model is then estimated using vector heterogeneous autoregressive (VHAR) models estimated with theleast absolute shrinkage and selection operator (LASSO). Our methodology  improves forecasting precision relative to standard benchmarks and leads to better estimates of the minimum variance portfolios.