Market dynamic is studied by quantifying the dependence of the entropy $S(\tau,n)$ of the clusters formed by the series of the prices $p_t$ and its moving average $\widetilde{p}_{t,n}$ on temporal horizon $M$. We report results of the analysis performed on high-frequency data of the Nasdaq Composite, Dow Jones Industrial Avg and Standard \& Poor 500 indexes downloaded from the Bloomberg terminal www.bloomberg.com/professional. Both raw and sampled data series have been analysed for a broad range of horizons $M$, varying from one to twelve months over the year 2018. A systematic dependence of the cluster entropy function $S(\tau,n)$ on the horizon $M$ has been evidenced in the analysed assets. Hence, the cluster entropy function is integrated over the cluster $\tau$ to yield a synthetic indicator of price evolution: the \emph{Market Dynamic Index} $I(M,n)$. Moreover, the \emph{Market Horizon Dependence} defined as $H(M,n)=I(M,n)-I(1,n)$ is calculated and compared with the values of the horizon dependence of the pricing kernel with different representative agent models obtained by a Kullback-Leibler entropy approach.
↧