Introduction - Hydrologic Time Series Analysis: Theory and Practice - page 6

a time series model and to evaluate goodness-of-fit of the developed model.
Finally, in the verification step, the developed time series model is evaluated
using independent sets of data. For further details of the time series analysis,
the readers are referred to the specialized topics on time series analysis such
as Yevjevich (1972), Salas et al. (1980), Bras and Rodriguez-Iturbe (1985),
Cryer (1986), and Clarke (1998).
1.4 Classification of Time Series
A time series can be classified in many ways according to different criteria.
Three widely used classifications of the time series are described below. The
details about the classification of hydrologic time series can be found in Salas
(1993).
1.4.1 Discrete or Continuous Time Series
Time series can be either continuous or discrete. A time series is called 'discrete'
if the observations are recorded at different time instants or at different points
in space (Haan, 1977; Shahin et al., 1993). On the other hand, if the observations
are recorded continuously in time or space, then the series is known as a
'continuous time series'. 'Discrete time series' is often derived from a
'continuous time series'. Usually in hydrology, a time series is of the discrete
type. As a result, the case studies presented in this topic are restricted to the
discrete time series. A continuous plot of a 'discrete time series' should not be
confused with a 'continuous time series'.
1.4.2 Full or Partial Duration Series
A 'full time series' is the one which contains all the recorded observations
over time or space (Haan, 1977; Shahin et al., 1993). As the name suggests,
a 'partial duration series' contains only selected observations which are
extracted from the full time series. For instance, daily rainfall recorded at a
specific location over a given period of time constitutes a full time series of
rainfall. A time series of one-day maximum rainfall can be extracted from the
full rainfall time series by arranging the maximum rainfall occurring in a day
for each year in the order of occurrence. Note that the maximum rainfall time
series contains less information than the original full rainfall time series. That
is, a 'partial duration series' always contains less information than the 'full
time series'. In addition, the observation points in a partial duration series
may not be equidistant.
1.4.3 Univariate or Multivariate
If only one variable is observed at each time, the time series is known as
'univariate time series'. However, if two variables are observed at the same
time (simultaneously), the series is known as 'bivariate time series'. If more