Sie sind auf Seite 1von 1

Source : Time series forecasting by Chattfield

1. A time-series is a collection of observations made sequentially through time


2. Applications of time-series forecasting include:
1. Economic planning
2. Sales forecasting
3. Inventory (or stock) control
4. Production and capacity planning
5. The evaluation of alternative economic strategies
6. Budgeting
7. Financial risk management
8. Model evaluation
Types of forecasting method:

Let we have an observed time series x1, x2 ,...,xN and wish to forecast
future values such as xN+h. The integer h is called the lead time or the
forecasting horizon (h for horizon) and the forecast of xN+h made at time
N for h steps ahead will be denoted by xN(h).
A forecasting method is a procedure for computing forecasts from present
and past values
Forecasting method may simply be an algorithmic rule and need not
depend on an underlying probability model. Alternatively it may arise from
identifying a particular model for the given data and finding optimal
forecasts conditional on that model. So method and model are clearly
different.
Forecasting methods may be broadly classified into three types:
(a) Judgemental forecasts: based on subjective judgement, intuition,
inside commercial knowledge, and any other relevant information.
(b) Univariate methods: where forecasts depend only on present and
past values of the single series being forecasted, possibly augmented by a
function of time such as a linear trend.
(c) Multivariate methods:where forecasts of a given variable depend, at
least partly, on values of one or more additional time series variables,
called predictor or explanatory variables. Multivariate forecasts may
depend on a multivariate model involving more than one equation if the
variables are jointly dependent.
More generally a forecasting method could combine more than one of the
above approaches

Das könnte Ihnen auch gefallen