Professional Documents
Culture Documents
Extrapolation:
- moving averages
- exponential smoothing (- both use special kinds of
averages of the most recent data to forecast)
- trend line analysis : (the comparision of regression
models of the rate of growth of data overtime) eg.
dependent variable – sales and independent variable-
f(t)
- the straight line projection (a linear trend is used)
- classical decomposition- assumes that data are made
up of at least 3 components (seasonality, trend and
randomness)
- method attempts to separate
- Box jetkins : a sophisticated statistical technique
which attempts to pick an optimal from a large no. of
posssibilities( detailed not required)
Causal
- Causal regression ( beyond the scope of the course)
(sales vs. funtion of advertisement and price)
- Simulation: develops a model of process and then
conducts a series of trial and error experiments to
predict the behaviour of process over time.
Judgemental forecasting:
3 types of seasonality
- Non seasonality
- Additive Seasonality
(seasonal fluctuation are of constant size)- less
common
- Multiplicative Seasonality
(seasonal fluctuations are proportional to the data)
The Naive Forecasting Model ( benchmark model)
Ft+1= Xt
Forecasting for next period = Observed value this period
et = x t – F t
error = Data – Forecast
-The mean error measures are computed only for the last half
of the data.
- The first part of data is used to fit the forecasting model
- Fitting data – Warm up sample
- second part – forecasting sample
Rule of thumb:
- to put at least six non-seasonal data points
- two complete seasons of seasonal data in warm up
sample
- If fewer data, no need to bother with two samples
In a long time series, common practice simply divide
half
MOVING AVERAGE FORECASTING
- UNWEIGHTED
(EQUAL WEIGHTS TO OLD AND NEW DATA)
- WEIGHTED
( MORE WEIGHTS TO MOST RECENT DATA)
Ft+1 = Ft + α et
Forecast fo t+1 = Forecast for t + α x Error in t
Here, α = Smoothing parameter ( 0 < α < 1)
At = α Xt + (1 – α) At-1
F t+1 = At Ft = At-1
Ft+1 = α Xt+ (1-α) Ft
After rearranging
Ft+1 = Ft + α et
New forecast = old forecast + a proportion of the error
between the observed demand and the old forecast
Ft+1 = a Xt + (1-a) Ft
Generalizing:
If α = 0.3, t=3
F4 = 0.3 X3 + 0.21X2+ 0.147X1 + 0.343 F1
Ft = a + bt
Disadvantages;
- All regression forecasts are based on a
single equation
- Re computing the changing trend is tideous
- Equal weight is assigned to all observations
- Even if model fits for warmup sample, it may
not be for later forecasting
S t = F t + α 1 et Tt= Tt-1+α 2 et
Ft+m= St + mTt
T Data Forecast Error Level at the end of t Trend at the end of t Forecast for t+1
(Xt) (Ft) et = X t - F t St= Ft + a1et Tt=Tt-1+a2et Ft+1 = St + Tt
0 S0= 54.9 T0= 1.7 F1=54.9+1.7= 56.6
1 60.0 56.6 3.4 S1= T1=1.7+.01(3.4)=1.7 F2=56.9+1.7=58.6
56.6+.1(3.4)=56.9
2 55.0 58.6 -3.6 S2= 58.6+.1(-3.6) T2=1.7+.01(-3.6) =1.7 F3=58.2+1.7=59.9
=58.2
3 64.0 59.9 4.1 S3=60.3 T3=1.7 F4=62.0
4 51.0 62.0 -11.0 S4=60.9 T4=1.6 F5=62.5
5 69.0 62.5 6.5 S5=63.2 T5=1.7 F6=64.9
6 66.0 64.9 1.1 S6=65.0 T6=1.7 F7=66.7
7 83.0 66.7 16.3 S7=68.3 T7=1.9 F8=70.2
8 90.0 70.2 19.8 S8=72.2 T8=2.1 F9=74.3
9 76.0 74.3 1.7 S9=74.5 T9=2.1 F10=76.6
10 95.0 76.6 18.4 S10=78.4 T10=2.3 F11=80.7
11 72.0 80.7 -8.7 S11=79.8 T11=2.2 F12=82.0
12 88.0 82.0 6.0 S12=82.6 T12=2.3 F13=84.9
13 84.9
ACTUAL DATA
DESEASONALIZED DATA = ------------------------------------------
SEASONAL INDEX
ACTUAL DATA
SEASONAL INDEX = ---------------------------------------------
AVERAGE FOR THE YEAR
IF DATA ARE MONTHLY, 12 SEASONAL INDICES
IF DATA ARE QUARTERLY, 4 SEASONAL INDICES