Quantitative forecasting methods in library management

1
Quantitative forecasting methods in library
management

• Prof. Dr. Algirdas Budrevicius
• Vilnius University, Faculty of Communication

Course website http//www.kf.vu.lt/albud/progn/E
ngl
2

• "If you can look into the seeds of time, and say
  which grain will grow and which will not, speak
  then unto me. "
• --William Shakespeare

3

• "It is far better to foresee even without
  certainty than not to foresee at all. "
• --Henri Poincare in The Foundations of Science,
  page 129.

4
Course plan

• Lecture 1. Forecasting history and current
  situation. Forecasting in management. Qualitative
  and quantitative forecasting. Time series
  forecasting. Visual data pattern analysis.
  Forecasting in library management. Naive
  forecasting methods.

5
Course plan (continued)

• Lecture 2. Part 1 Moving average forecasting
  method. Errors of forecast. Part 2 Practical
  work with Excel
• Lecture 3. Part 1 Forecasting using linear
  regression. Trend analysis. Part 2 Practical
  work with Excel
• Lecture 4-5. Forecasting project analysis of
  forecasting situations in libraries examples.
  Practical work with Excel
• Lecture 6. Discussions

6
Course materials

• Course description Website http//www.kf.vu.lt/a
  lbud/progn/Engl
• Lectures PowerPoint presentations
• Data, demonstrations, task solutions Excel
  workbooks

7
Development of the forecasting technique

• Non scientiffic forecasting e.g. Astrology, Book
  of Changes.
• 19-20 century. Demographic forecasts
• Development of the quantitative methods
  middle-to-second part of the 20th century.
• New developments Neural network based methods

8
Current situation in forecasting

• Forecasting is widely used in management now
• There exist a well defined set of quantitative
  forecasting methods that changes very little
  during last fiew decades
• There exists computer software that may be quite
  simply applied in forecasting
• Excel program allows to solve simple forecasting
  tasks

9
Forecasting in management
Forecasting is used in various domains of
management, such as

• Personnel management
• Resource management
• Finance management
• Organizational management

10
Taxonomy of forecasting methods

• Methods quantitative and qualitative
• Qualitative judgmental (based on expert
  opinions) and technological (used for long term
  forecasts)
• Quantitative time series methods and reasoning
• Note only time series methods will be considered
  in this course.

11
Definition of a forecasting situation

• Data (time series, or historical data)
• Forecasting method (e.g. Moving average, Trend
  analysis)
• Forecast
• Error of forecast

12
Quantitative time series based forecasting
13
Naive forecasts NF1 and NF2

• Naive forecasts (a folk forecasting technique)
  
• NF1. (The value tomorow will be the same as
  today). Example Number of library visitors
  today was 120. Forecast NF1 for tomorow 120.
• NF2. (The value tomorow will be less (greater)
  by 10 ). Example Average temperature this
  month is 20 degrees. Forecast NF2 for the next
  month Temperature will be 25 degrees (increase
  of 25).

14
Time-series methods of forecasting

• Time series analysis relies on historical data
  and attempts to project historical patterns into
  the future
• Note only time series methods will further be
  considered

15
Time-series example

• Number of visitors in a library (in th.)

16
Recomended form to present data and forecasts an
example
17
Example of real time series data concerning
libraries

• Number of libraries (network)
• Document stocks
• Loan of documents
• Number of users
• Number of visitors, etc. (also see examples in
  Excell worksheets)

Conclusion good possibilities to apply
forecasting methods, based on time series analysis
18
Example of data
19
Example of forecasting
20
Patterns of the time-series data
A forecasting method should comply with the data
pattern. There are 4 basic data patterns

• Horizontal (random, irregular variation)
• Trend (linear)
• Periodical (cyclical, seasonal)
• Complex (a combination of part or all listed
  above)

21
Horizontal pattern
22
Trend
23
Periodical pattern
24
Complex pattern
25
Measuring forecast accuracy

• What is the accuracy of a particular forecast?
• How to measure the suitability of a particular
  forecasting method for a given data set?

26
Definition of the forecast error

• Error (e) of a forecast is measured as a
  difference between the actual (A) and forecasted
  values (F), that is,
• eA-F,
• or, in a relative form e100 (A-F)/A.
• The error can be determined only when actual
  (future) data are available.

27
Standard statistical measures to estimate errors
(1)

• To preliminary evaluate a forecast and
  suitability of a method, various statistical
  measures may be used. In evaluating forecasts
  obtained by means of the moving average method,
  the following measures may be used

• Mean (average) error (ME)
• Mean absolute error (MAE)
• Mean squared error (MSE)

28
Standard statistical measures to estimate errors
(2 - relative)

• Mean percentage error (MPE)
• Mean absolute percentage error (MAPE)

29
Statistical measures of goodness of fit
In trend analysis the following measures will be
used

• The Correlation Coefficient
• The Determination Coefficient

30
The Correlation Coefficient

• The correlation coefficient, R, measure the
  strength and direction of linear relationships
  between two variables. It has a value between 1
  and 1
• A correlation near zero indicates little linear
  relationship, and a correlation near one
  indicates a strong linear relationship between
  the two variables

31
The Coefficient of Determination

• The coefficient of determination, R2, measures
  the percentage of variaion in the dependent
  variable that is explained by the regression or
  trend line. It has a value between zero and one,
  with a high value indicating a good fit.

32
End