1: \begin{abstract}

2: 	The conditional variance, skewness, and kurtosis play a central role in time series analysis. These three conditional moments (CMs)

3: are often studied by some parametric models but with two big issues: the risk of model mis-specification and the instability of model estimation.

4: To avoid the above two issues, this paper proposes a novel method to estimate these three CMs by the so-called quantiled CMs (QCMs).

5: The QCM method first adopts the idea of Cornish-Fisher expansion to construct a linear regression model, based on $n$ different estimated conditional quantiles. Next, it computes the QCMs simply and simultaneously by using the ordinary least squares estimator of

6: this regression model, without any prior estimation of the conditional mean.

7: Under certain conditions, the QCMs are shown to be consistent with the convergence rate $n^{-1/2}$. Simulation studies indicate that the QCMs perform well under different scenarios of Cornish-Fisher expansion errors and quantile estimation errors.

8: In the application, the study of QCMs for three exchange rates demonstrates

9: the effectiveness of financial rescue plans during the COVID-19 pandemic outbreak, and suggests that the existing  ``news impact curve'' functions for the conditional skewness and kurtosis may not be suitable.

10: \end{abstract}

11: