Introduction to ArtificiaI Intelligence in Higher Education
Forecasting
1. Forecasting and Methods
of Forecasting
Ashish Kumar
Center for Tourism Management
Master in Travel and Tourism Management
Dev Sanskriti Vishwavidyalaya
Gayatrikunj-Shantikunj
Haridwar
2. Forecasting is a planning tool used to
predict future environmental
happenings that will influence the
operations of an organisation.
Forecasting
3.
Forecasting is the essence of planning.
Forecasting forces managers to look ahead
Forecasting helps in achieving better coordination
Forecasting, by revealing the weak spots in the organisation
Effective forecasting helps in identifying the environmental forces
Advantages of Forecasting
4. There are three types of forecasting
1. Judgmental or Qualitative methods
2. Extrapolative or Time series methods
3. Causal or Explanatory methods
Types of Forecasting
5. * Judgmental forecasting methods incorporate intuitive judgment,
opinions and subjective probability estimates.
* Judgmental forecasting is used in cases where there is lack of
past or historical data or during completely new and unique
market conditions.
* Useful for medium to long range forecasting tasks.
* Provide a basis for some important decisions.
Judgmental Forecasting Methods
6. Three important Judgmental methods are :-
Delphi Technique - Develop forecast through group
consensus.
Market Surveys - Involves the use of questionnaires,
consumer panels & tests of new products & services.
Scenario Writing - process of analyzing possible future
events by considering alternative possible outcomes.
Judgmental Forecasting Methods
7. Use past history of demand.
Comprised of four separate components: trend
component, cyclical component, seasonal component,
and irregular component.
The objective of this method is to identify the pattern in
historic data & extrapolate this for future.
Extrapolative or Time series Methods
8. Moving Average Method - average of demands
occurring in several of the most recent periods.
Weighted Moving Average - allows for varying
weighting of old demands.
Exponential Smoothing – exponentially decreases the
weighting of old demands.
Types of Extrapolative Method
9. A statistical forecasting model based
on historical demand data as well as
on variables believed to influence
demand.
Causal Forecasting
10.
Regression analysis - a functional relationship is
established between variables from the historical data
and then used to forecast dependent variable values.
Econometric method – an extension of regression
analysis and include a system of simultaneous
regression equations.
Types of Causal Forecasting
11. Reference
VSP Rao, V Hari Krishna, (2007) Management text and cases, Excel
Books publication, page – 131-135
Rob J Hyndman, George Athanasopoulos, (2014) Forecasting -
principle and practice, Otext, page – 25-62