Invited review abstract

Coronal mass ejections (CMEs) have long been
recognized as a major manifestation of solar
activity, and as a vital agent for the interplanetary
propagation of solar disturbances. Attempts to
understand their nascency and post-eruptive
development were greatly intensified in recent years,
particularly in the framework of space weather
forecasting. Despite these efforts, predicting their
IP evolution remains a challenging task due to a
CME's elaborate structure and its delicate
interaction with both the ambient magnetized medium
and other CMEs. This talk will summarize past and
recent attempts to predict a CME's trajectory, travel
time, and geo-effectiveness using both observational
statistics and theoretical modelling. Except for few
analytical approaches, the latter usually employs
numerical MHD simulations, either as "principal"
(employing idealized settings with few control
parameters) or "realistic" (aiming to reproduce data
from actual events) studies. Selected results from
both categories will be discussed, along with their
predictive strengths and possible shortcomings.