Unformatted text preview: ES carries all past history. MA eliminates “bad” data
after N periods
after
MA requires all N past data points while ES only
requires last forecast and last observation.
requires 12 Regression Analysis
Regression Methods can be used when a trend is
present. Let (x1, y1)…. be n paired data points for two
present.
)….
variables which x iis the independent variable and y iis the
s
s
dependent one (on x). ˆ
y = a + bx ˆ
y is the predicted value of the response y. In forecasting
problems, usually the independent variable x iis time, t.
s
We have to estimate the values of a and b.
and Regression Analysis b= N ∑ ty − ∑ t ∑ y
2 N ∑ t − (∑ t ) a= 2 ∑ y − b∑ t
N 13 Example
Year
1
2
3
4
5
6
7
8
9
10 Sales
1000
1300
1800
2000
2000
2000
2200
2600
2900
3200 Seasonality
Seasonality corresponds to a pattern in the data that
repeats at regular intervals.
repeats
There may be multiple seasonal factors: c1 , c2 , . . . , cN
There
where i = 1 is first period of season, i = 2 is second
where
period of the season, etc..
period
ci = 1.25 implies 25% higher than the baseline on avg.
ci = 0.75 implies 25% lower than the baseline on avg. 14 Seasonality Illustration Method of Estimating Seasonal
Method
Factors
Factors
Compute the sample mean of the entire data set (should
be at least several seasons of data).
be
Divide each observation by the sample mean. (This
gives a factor for each observation.)
gives
Average the factors for like periods in a season.
Average
The resulting N numbers will ex...
View
Full Document
 Spring '14
 Forecasting, Regression Analysis, Time series analysis

Click to edit the document details