Learning objectives:
At the end of this session,
1. the participant should
understand forecasting methods.
2. the participant should know
concepts behind forecasting models.
Performance objectives:
1. the participant should be able
to develop times series model for
forecasting epidemic.
I. Background:
Unaided, subjective judgements to warn
of forthcoming events and changes are not as
accurate and effective as systematic, explicit
approaches to forecasting.
This does not mean there is error free
forecasts. This does mean explicit systematic
forecasting approaching can provide
substantial benefits when used properly as all
types and forms of forecasting techniques are
made available within the existing data.
1. Judgmental method:
Forecasts are made as
individual judgements or by
committee agreement or
decisions.
2. Quantitative method:
To know what will happen, but not why
something happens.
There are three subcategories of this method
2.1 Times series methods: Seek to identify
historical patterns (using time as a reference)
and then forecast using a time-based
extrapolation of those patterns.
3. Technological method:
Address long-term issues of a
technological, societal, political or
economic nature.
3.1 Extrapolative methods: using historical
patterns and relationships as a basic for forecasts
3.2 Analogy-based methods: using historical
and other analogies to make forecasts
3.3 Expert-based methods:
3.4 Normative-based methods: {using
objectives, goals, and desired outcomes as a
basic for forecasting, thereby influencing future
events}.
Additive model
1. We assume that the data is the sum of the
time series components.
Multiplicative model
1. We assume that the data is
the product of the various components.
Yt = Trt * Snt * t
3. The seasonal factor of multiplicative model is a
proportion (ratio) to the trends, and thus the
magnitude of the seasonal swing increases or
decreases according to the behaviour of trend.
Objectives of modelling :
1) to monitor the malaria situation
in the study area and forecast with
modelling;
Methods:
1. This is a documentary study using time series
data covering 1984 to 1992.
2. The dependent variable was the incidence of
malaria occurring during a given time including both
out-patient and in-patient malaria cases.
3. For a starting point, we demonstrated a simple,
two-variable regression model using the
Results:
The output for MINITAB program illustrating
seasonal indices and centred moving average.
Times series (multiplicative decomposition
method)
Seasonal Indices
Period Index
1
1.18483
2
0.309150
3
0.738706
4
1.76732
Accuracy of Model
MAPE:
494
MAD:
234
MSD:
101789
Actual
Predicted
Forecast
Actual
Predicted
Forecast
Cases
1000
500
MAPE:
MAD:
MSD:
0
0
10
20
30
40
249.4
206.1
89899.9
1500
Actual
Predicted
Actual
Predicted
cases
1000
Moving Average
500
0
0
10
20
30
Length:
MAPE:
499
MAD:
256
MSD:
132728
Coef
21.1
StDev
110.5
T
0.19
P
0.850
10.932
5.209
2.10
0.043
R-Sq = 11.5%
R-Sq(adj) = 8.9%
1000
se
sa
C
500
0
Regression
95% CI
-500
95% PI
0
10
20
quarterlytime periods
30
40
Discussion:
1.Epidemic of malaria: What?
1.1 Periodical rapid and great increase in
malaria morbidity and perhaps mortality, reaching
levels above local average endemicity.
1.2 A rapid increase in malaria morbidity and
mortality in a given population (independent of
Points to ponder:
1. Among diverse factors, the selection of
independent variables should be judiciously
based on theoretical considerations.
2. It is worth emphasizing that the simple, twovariable regression model is limited in
information.
3. The preferred approach is to perform
A cautionary note:
References:
1. Armitage P, Berry G. Automatic selection
procedures and colinearity. In:Statistical Methods
in Medical Research. 3rd ed. Blackwell Scientific
Publications, Oxford. 1994; 321-323.
References (cont):
4. Foster DP, Stine RA, Waterman RP. Summary
regression case. In:Business Analysis Using
Regression: A Case Book. Springer-Verlag New York,
Inc. 1998; 227.
5. Gujarati DN. Test of specification errors. In: Basic
Econometric. 3rd ed. Mcgraw-Hill, Inc. Singapore.