This document provides an overview of time series forecasting techniques. It discusses the components of time series data including trends, cycles, seasonality and irregular fluctuations. It also covers stationary and non-stationary time series. Forecasting techniques covered include naive methods, smoothing techniques like moving averages and exponential smoothing, and decomposition methods. Regression models for trend analysis and measuring forecast accuracy are also discussed.
1. Time Series Forecasting: Objectives
• Gain a general understanding of time series forecasting
techniques.
• Understand the four possible components of time-series
data.
• Understand stationary forecasting techniques.
• Understand how to use regression models for trend analysis.
• Learn how to decompose time-series data into their various
elements and to forecast by using decomposition
techniques.
• Understand the nature of autocorrelation and how to
test for it.
• Understand auto-regression in forecasting.
2. Time-Series Forecasting
• Time-series data: data gathered on a given
characteristic over a period of time at regular
intervals
• Time-series techniques
o Attempt to account for changes over time by
examining patterns, cycles, trends, or using
information about previous time periods
o Naive Methods
o Averaging
o Smoothing
o Decomposition of time series data
3. Time Series Components
• Trend – long term general direction, typically 8 to
10 years
• Cycles (Cyclical effects) – patterns of highs and
lows through which data move over time periods
usually of more than a year, typically 3 to 5 years
• Seasonal effects – shorter cycles, which usually
occur in time periods of less than one year.
• Irregular fluctuations – rapid changes or “bleeps”
in the data, which occur in even shorter time
frames than seasonal effects.
5. Time Series Components
• Stationary time-series - data that contain no trend,
cyclical, or seasonal effects.
• Error of individual forecast et – the difference
between the actual value xt and the forecast of that
value Ft i.e.
6. Measurement of Forecasting Error
• Error of the Individual Forecast (et = Xt – Ft) is the
difference between the actual value xt and the forecast of
that value Ft.
• Mean Absolute Deviation (MAD) - is the mean, or
average, of the absolute values of the errors.
• Mean Square Error (MSE) - circumvents the problem
of the canceling effects of positive and negative forecast
errors.
– Computed by squaring each error and averaging the
squared errors.
7. Measurement of Forecasting Error
• Mean Percentage Error (MPE) – average of the
percentage errors of a forecast
• Mean Absolute Percentage Error (MAPE) – average of
the absolute values of the percentage errors of a
forecast
• Mean Error (ME) – average of all the errors of
forecast for a group of data
11. Smoothing Techniques
• Smoothing techniques produce forecasts based on
“smoothing out” the irregular fluctuation effects in
the time-series data
• Naive Forecasting Models - simple models in which
it is assumed that the more recent time periods of
data represent the best predictions or forecasts for
future outcomes
12. Smoothing Techniques
• Averaging Models - the forecast for time period
t is the average of the values for a given number of
previous time periods:
o Simple Averages
o Moving Averages
o Weighted Moving Averages
• Exponential Smoothing - is used to weight data
from previous time periods with exponentially
decreasing importance in the forecast.
13. Simple Average Model
The forecast for time
period t is the average of
the values for a given
number of previous time
periods.
Month Year
Cents
per
Gallon Month Year
Cents
per
Gallon
January 2 61.3 January 3 58.2
February 63.3 February 58.3
March 62.1 March 57.7
April 59.8 April 56.7
May 58.4 May 56.8
June 57.6 June 55.5
July 55.7 July 53.8
August 55.1 August 52.8
September 55.7 September
October 56.7 October
November 57.2 November
December 58.0 December
The monthly average last
12 months was 56.45,
so I predict
56.45 for September.
14. Moving Average
• Updated (recomputed) for every new time period
• May be difficult to choose optimal number of periods
• May not adjust for trend, cyclical, or seasonal effects
Update each period.
15. Demonstration Problem 15.1:
Four-Month Moving Average
Shown in the following table here are shipments
(in millions of dollars) for electric lighting and wiring
equipment over a 12-month period. Use these data
to compute a 4-month moving average for all
available months.
16. Demonstration Problem 15.1:
Four-Month Moving Average
Months Shipments
4-Mo
Moving
Average
Forecast
Error
January 1056
February 1345
March 1381
April 1191
May 1259 1243.25 15.75
June 1361 1294.00 67.00
July 1110 1298.00 -188.00
August 1334 1230.25 103.75
September 1416 1266.00 150.00
October 1282 1305.25 -23.25
November 1341 1285.50 55.50
December 1382 1343.25 38.75
18. Weighted Moving Average
Forecasting Model
A moving average in which some time periods are
weighted differently than others.
Example of 3 months
Weighted average
where last month’s value
value for the previous month
value for the month before the
previous month
The denominator = the total of weights
1tM
2tM
3tM
19. Demonstration Problem 15.2:
Four-Month Weighted Moving Average
Months Shipments
4-Month
Weighted
Moving
Average
Forecast
Error
January 1056
February 1345
March 1381
April 1191
May 1259 1240.88 18.13
June 1361 1268.00 93.00
July 1110 1316.75 -206.75
August 1334 1201.50 132.50
September 1416 1272.00 144.00
October 1282 1350.38 -68.38
November 1341 1300.50 40.50
December 1382 1334.75 47.25
20. Exponential Smoothing
Used to weight data from previous time periods with
exponentially decreasing importance in the forecast
t t t
t
t
t
F X F
F
F
X
where
1
1
1
: the forecast for the next time period (t+1)
the forecast for the present time period (t)
the actual value for the present time period
= a value between 0 and 1
is the exponential
smoothing constant
21. Demonstration Problem 15.3: = 0.2
The U.S. Census Bureau reports the total units of new
privately owned housing started over a 16-year recent
period in the United States are given here. Use
exponential smoothing to forecast the values for each
ensuing time period. Work the problem using = 0.2,
0.5, and 0.8
24. Trend Analysis
• Trend – long run general direction of climate over
an extended time
• Linear Trend
• Quadratic Trend
• Holt’s Two Parameter Exponential Smoothing -
Holt’s technique uses weights (β) to smooth the
trend in a manner similar to the smoothing used in
single exponential smoothing (α)
25. Average Hours Worked per Week
by Canadian Manufacturing Workers
Following table provides the data needed to
compute a quadratic regression trend model on
the manufacturing workweek data
26. Average Hours Worked per Week
by Canadian Manufacturing Workers
Period Hours Period Hours Period Hours Period Hours
1 37.2 11 36.9 21 35.6 31 35.7
2 37.0 12 36.7 22 35.2 32 35.5
3 37.4 13 36.7 23 34.8 33 35.6
4 37.5 14 36.5 24 35.3 34 36.3
5 37.7 15 36.3 25 35.6 35 36.5
6 37.7 16 35.9 26 35.6
7 37.4 17 35.8 27 35.6
8 37.2 18 35.9 28 35.9
9 37.3 19 36.0 29 36.0
10 37.2 20 35.7 30 35.7
27. Excel Regression Output using
Linear Trend
Regression Statistics
Multiple R
R Square
Adjusted R Square
Standard Error
Observations
ANOVA
SS MS F Significance F
Regression 1 13.4467 13.4467 51.91 .00000003
Residual 33 8.5487 0.2591
Total 34 21.9954
Coefficients Standard Error t Stat P-value
Intercept 37.4161 0.17582 212.81 .0000000
Period -0.0614 0.00852 -7.20 .00000003
i ti i
t
Y X
X
where
Y
0 1
37 416 0 0614
:
. .
data value for period i
time period
i
i
Y
X
35
0.782
0.611
0.5600
0.509
df
28. Excel Regression Output using
Quadratic Trend
Regression Statistics
Multiple R 0.8723
R Square 0.761
Adjusted R Square 0.747
Standard Error 0.405
Observations 35
ANOVA
df SS MS F Significance F
Regression 2 16.7483 8.3741 51.07 1.10021E-10
Residual 32 5.2472 0.1640
Total 34 21.9954
Coefficients Standard Error t Stat P-value
Intercept 38.16442 0.21766 175.34 2.61E-49
Period -0.18272 0.02788 -6.55 2.21E-07
Period2 0.00337 0.00075 4.49 8.76E-05
i ti ti i
ti
t t
Y X X
X
X X
where
Y
0 1 2
2
2
2
38164 0183 0 003
:
. . .
data value for period i
time period
the square of the i period
i
i
th
Y
X
30. Demonstration Problem 15.4
Data on the employed U.S. civilian labour force (in
100,000) for 1991 through 2008, obtained from the U.S.
Bureau of Labor Statistics. Use regression analysis to fit
a trend line through the data and explore a quadratic
trend. Compare the models.
33. Time Series: Decomposition
Decomposition – Breaking down the effects of
time series data into four component parts:
trend, cyclical, seasonal, and irregular
Basis for analysis is the Multiplicative Model
Y = T · C · S · I
where:
T = trend component
C = cyclical component
S = seasonal component
I = irregular component
34. Household Appliance Shipment Data
Illustration of decomposition process: the 5-year quarterly
time-series data on U.S. shipments of household appliances
Year Quarter Shipments Year Quarter Shipments
1 1 4009 4 1 4595
2 4321 2 4799
3 4224 3 4417
4 3944 4 4258
2 1 4123 5 1 4245
2 4522 2 4900
3 4657 3 4585
4 4030 4 4533
3 1 4493
2 4806
3 4551
4 4485
Shipments in $1,000,000.
37. Eliminate the Max and Min
for each Quarter
1 2 3 4 5
Q1 96.84% 100.22% 100.09% 94.84%
Q2 104.62% 106.17% 105.57% 108.13%
Q3 102.06% 106.34% 99.01% 98.74%
Q4 94.40% 90.34% 97.32% 95.85%
Eliminate the maximum and the minimum for each quarter to
remove irregular fluctuations. Average the remaining ratios for
each quarter.
38. Computation of Average
of Seasonal Indexes
1 2 3 4 5 Average
Q1 96.84% 100.09% 98.47%
Q2 106.17% 105.57% 105.87%
Q3 102.06% 99.01% 100.53%
Q4 94.40% 95.85% 95.13%
40. Autocorrelation (Serial Correlation)
• Autocorrelation occurs in data when the error terms of
a regression forecasting model are correlated and not
independent, particularly with economic variables.
• Potential Problems
• Estimates of the regression coefficients no longer have
the minimum variance property and may be inefficient.
• The variance of the error terms may be greatly
underestimated by the mean square error value.
• The true standard deviation of the estimated regression
coefficient may be seriously underestimated.
• The confidence intervals and tests using the t and F
distributions are no longer strictly applicable.
43. Durbin-Watson Test
H
Ha
0 0
0
:
:
D
t t
where
e e
e
t
n
t
t
n
2
2
2
1
1
: n = the number of observations
If D > do not reject H (there is no significant autocorrelation).
If D < , reject H (there is significant autocorrelation).
If , the test is inconclusive.
U
0
L
0
L U
d
d
d d
,
D
45. Overcoming Autocorrelation Problem
• Addition of Independent Variables
• Transforming Variables
First-differences approach - Often autocorrelation occurs
in regression analyses when one or more predictor
variables have been left out of the analysis
Percentage change from period to period - each value of x
is subtracted from each succeeding time period value of x;
these “differences” become the new and transformed x
variable, the same for y
Use autoregression - multiple regression technique in
which the independent variables are time-lagged versions
of the dependent variable