Testing the maximal rank of the volatility process for continuous diffusions observed with noise (Q2405147)

The paper puts forward a test statistic for the maximal rank of the stochastic volatility process in continuous diffusion models exhibiting noise. An example of such an application is the financial data that is exposed to microstructure noise at ultra-high frequencies. In this respect, the article gives a methodology that combines a matrix perturbation method with the pre-averaging technique and shows the asymptotic mixed normality of the proposed test statistic. A real data scenario of financial data regarding stocks of eight American banks is also investigated to show the practical side of the theoretical contribution.