A useful feature may be to determine the normal frequency of round-sized trades, and monitor deviations from that expected value. The ML algorithm could, for example, determine if a larger-than-usual proportion of round-sized trades is associated with trends, as human traders tend to bet with a fundamental view, belief, or conviction. Conversely, a lower-than-usual proportion of round-sized trades may increase the likelihood that prices will move sideways, as silicon traders do not typically hold long-term views.

---

This indicator is based on Advances in Financial Machine Learning 2018 - Marcos Lopez de Prado book, Page 293.

nclick="viewGallery_GalleryManager.GoToPrev();"> nclick="viewGallery_GalleryManager.GoToPrev();">

ta itemprop="image" content="https://c.mql5.com/31/325/order-size-interval-screen-6520.png" />

nclick="viewGallery_GalleryManager.GoTonext();"> nclick="viewGallery_GalleryManager.GoTonext();">