From a stochastic point of view, the problem of forecasting
future values of a random variable can be seen as the determination
of the probability density function of future values conditioned
by past observations [9]. Yevjevich as reported in [10]
was among the first at attempting a prediction of properties of
droughts using the geometric probability distribution, defining
a drought of k years as k consecutive years when there are no
adequate water resources. Rao and Padmanabhan [11] investigated
the stochastic nature of yearly and monthly Palmer’s
drought index (PDI) series to characterize them via valid
stochastic models which may be used to forecast and to
simulate the PDI series. Sen [12] derived exact probability
distribution functions of critical droughts in stationary second
order Markov chains for finite sample lengths on the basis of
the enumeration technique and predicted the possible critical
drought durations that may result from any hydrological
phenomenon. Lohani and Loganathan [13] used PDI in a
non-homogenous Markov chain model to characterize the
stochastic behavior of drought and based on these drought
characterizations an early warning system was used for
drought management [14].