ENGINYERIA EN ORGANITZACIÓ INDUSTRIAL

26503 ESTADÍSTICA INDUSTRIAL

COMPUTER LAB ACTIVITIES (MINITAB)

TIME SERIES ANALYSIS: Introduction

1.

-Quantitative forecasting methods characteristics:

-Use of statistical tools and historical data to predict the future.

-Used when past information about the variable being forecast is available.

-The information can be quantified (numerical facts).

-It can be assumed that the pattern of the past will appear in the future.

-Are more used to undertake long-term forecasting.

-Can be developed using:

- Time series methods: The historical data are restricted to past values of the
variable. The objective is to discover a pattern in the historical data and then extrapolate the
pattern into the future

- Causal forecasting methods: Based on the assumption that the variable we are
forecasting has a cause-effect relationship with one or more other variables.

-Quanlitative forecasting methods characteristics:

-Generally, involve the use of expert judgment to develop forecasts based on

their experience and instincts and include interpretation of data.

-Is often employed where the key trends or developments are hard to capture or
products where no historical information is available.

-Are preferred to undertake short-term forecasting.

-Based on surveys, interviews and opinions.

-Include the use of information gathered from Expert opinion, Market research,
Focus groups, Historical analogy, Delphi method and Panel consensus.

2. Describe each of the components in a time series.

Time series patterns can be usually described by 2 basic components:

-Trend: The increasing or decreasing value in the series.

Represents a general systematic component that changes over time
and does not repeat.

-Seasonality: It appears when there is a certain behaviour

(cycles) that repeats itself in systematic intervals over time.

1
 3. What does an autocorrelation coefficient measure?

Autocorrelation is a statistical method used for time series analysis whose purpose is to measure the
correlation of two values in the same data set at different time steps. It can detect non-randomness in a
data set. If the values in the data set are not random, then autocorrelation can help the analyst chose an
appropriate time series model.

4. Each of the following statements describes either a stationary or nonstationary series.

Indicate which:
a. A series that has a trend
Non-Stationary Time Series.

b. A series whose mean and variance remain constant over time

-Stationary Time Series.

c. A series whose mean value is changing over time

-Non-Stationary Time Series.

d. A series that contains no growth or decline

- Stationary Time Series.

5. List some of the forecasting techniques that should be considered when forecasting a
stationary series. Give examples of situations in which these techniques would be
applicable.
-Stationary Time Series are those that have trend and
multiplicative seasonality. For these methods, Classical
Decomposition and Winter´s method can be used.

6. List some of the forecasting techniques that should be considered when forecasting a
seasonal series. Give examples of situations in which these techniques would be applicable.
-Tend decomposition, Moving Averages, Single Exponential and
Double Exponential smoothing do not support seasonal patterns. For
seasonal patterns Triple Exponential and Classical Decomposition can
be used.

7. The number of marriages in Europe is given below. Compute the first differences for these data.
Plot the original data and the difference data as a time series. Is there a trend in either of these
series? Discuss.

Yea Marriages Yea Marriages

r r
1994 2413 2001 2362
1995 2407 2002 2334
1996 2403 2003 2362
1997 2396 2004 2336
1998 2403 2005 2344
1999 2443 2006 2384
2000 2371 2007 2244

First, a difference column. It is done in Statistics->Time Series-

>Differences

2
 Il·lustració 1. Time Series Worksheet including Differences Column.

Secondly, both columns can be plotted.

Il·lustració 2. Marriages plot.

Il·lustració 3.Differences plot.

By analysing the plots and using the tool Statistics->Time Series-

>Trend Analysis:

3
 Il·lustració Trend Analysis for Marriages.

Il·lustració 4. Trend Analysis for Differences.

8. Which measure of forecast accuracy should be used in each of the following situations?
a. The analyst feels that the size, or magnitude, of the forecast variable is important
in evaluating the accuracy of the forecast

-MAD, because it expresses the accuracy with the same units

as the original Time Series.

b. The analyst needs to penalize large forecasting errors

-As seen in class, the Mean Squared Deviation (MSD) penalizes large
fitting errors because errors are squared.

4
9. Alice Vila, the chief loan officer of European Bank, would like to analyze the bank’s loan
portfolio for the years 2002 to 2007. The data are shown below.
a. Compute the autocorrelation for time lags 1 and 2. Test to determine whether
these autocorrelation coefficients are significantly different from zero at the 0.05
significance level

Ilustración 1. Autocorrelation for legs 1 and 2

b. Plot the data and compute the autocorrelations for the first six time lags. Is this
time series stationary?

Yea Mar. 31 June 30 Sep. 30 Dec.

r 31
2002 2313 2495 2609 2792
2003 2860 3099 3202 3161
2004 3399 3471 3545 3851
2005 4458 4850 5093 5318
2006 5756 6013 6158 6289
2007 6369 6568 6646 6861

Ilustración 2. Portpoflio Graph.

5
 Ilustración 3. Autocorrelation until lag 6.

-Since it is evident that the time series has trend and the
autocorrelation is different to 0, I would say that it is stationary.
Unfortunately, I can not decide which is the length of the seasonal pattern,
since there are many lags where the value is higher than 0.

10. File TSA1_1.MTW contains the weekly sales of a food item for 52 consecutive weeks.
a. Plot the sales data as a time series
After having opened the Minitab Worksheet, it can be plotted:

b. Do you think this series is stationary or nonstationary?

At first sight, I would say that the time series is stationary due
to a possible seasonality.

6
 Il·lustració 5. Trend Analysis.

c. Compute the autocorrelations of the sales series for the first 10 time lags. Is the
behaviour of the autocorrelation consistent with your choice in part b? Explain

Ilustración 4. Autocorrelation.

-Yes, it is consistent with the first answer. It can be seen that an

autocorrelation of order 9 is present in the time series.