Scribd
Upload
116 views21 pages

Forecasting

This document provides an overview of quantitative forecasting methods, including time series analysis, simple moving averages, and exponential smoothing. It defines key components of time series such as trend, seasonality, cyclical patterns, and randomness. Simple moving average and exponential smoothing methods are explained as techniques to smooth fluctuations in time series data and generate forecasts. The document concludes that moving averages and exponential smoothing are commonly used forecasting models that smooth historical data to make single-period-ahead predictions.

Uploaded by

jorielle
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
116 views21 pages

Forecasting

This document provides an overview of quantitative forecasting methods, including time series analysis, simple moving averages, and exponential smoothing. It defines key components of time series such as trend, seasonality, cyclical patterns, and randomness. Simple moving average and exponential smoothing methods are explained as techniques to smooth fluctuations in time series data and generate forecasts. The document concludes that moving averages and exponential smoothing are commonly used forecasting models that smooth historical data to make single-period-ahead predictions.

Uploaded by

jorielle
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 21

Forecasting

Part 1
Presented by group 5
Topics
Quantitative Methods
01 and Characteristics
Time-series
02 Analysis
Simple Moving
03 Average

04 Exponential Smoothing
Introduction
● Modern society utilizes forecasting as one of the methods of knowledge.
Forecasting has many applications in the economy and industries, particularly in
management. The significance of solving forecasting problems depends on the
degree of risks reduction in making decisions, such as financial planning, inventory
control production scheduling, manufacturing processes, and investment strategies.

● Forecasting is a statistical task for making better decisions. The term, however, is
frequently confused with goals and planning. Forecasting is about predicting all
given information available as accurate as possible, including historical data and
knowledge of any future events that might impact the forecast. Goals are things that
you intend to happen and should be linked to forecasts and plans. Planning is done
after making a forecast and goals have been established.
01
Quantitative
Methods and
Characteristics
To a great extent, the suitability of forecasting method depends on the available data. If
there is a scarcity in reliable data, then qualitative forecasting methods are appropriate. Such
methods do not imply that they will be purely utilizing "spur of the moment guesses. There
are well developed structured approaches that can be used in absence of historical data.

Quantitative forecasting methods are applicable under two conditions:

(a) Availability of past numerical information.


(b) Past patterns can be assumed to continue into the future.

Several quantitative methods have been developed with specific purposes over the years. In
choosing a method, one must consider the properties, accuracies, and cost of application.
The most well-known methods are those used involving time-series data (collected at
regular intervals over time) and cross-sectional data (collected at a single point in time).
02
Time-series
Analysis
Time-series Analysis
According to otexts.com, a time series can be through as a list of numbers,
along with some information about what times those numbers were recorded.
To describe time series, the terms such as "trend," "seasonal," "random," and
"cyclical" are to be defined carefully.
TREND

It is the long run direction of the series


that includes any constant value in the
data. It could be in a shape of straight
line or linear trend that shows a CYCLICAL
constant amount of change. An
example would be population growth. A cycle occurs when the data exhibits
rise and falls that are not based on a
fixed frequency. These fluctuations are
usually due to economic conditions or
the so-called "business cycle." This is
only obvious in time series that span
several years. For example, economic
activities change in each country after
several years.
SEASONAL

Forecasting factors are based primarily


on nature and also on human behavior.
For example, seasonal demands
include demands for certain goods like
swimsuits, fur coats, gift items, etc.

RANDOM

These are events or effects that cannot


be predicted with certainly as they do
not exhibit any pattern. An example
would be an increase in food prices
brought about by a pandemic.
Simple Moving
03 Average
SIMPLE MOVING AVERAGE
Simple moving average is the unweighted average of a consecutive numbers of data
point. This forecasting method uses historical data as a means to eliminate the effects
of seasonal, cyclical, and erratic fluctuations. Simple moving average can be
calculated by choosing the number of items in time-series data to include in the
average. For each time period changes, a new data is added, and the old-time period
data is removed before calculating a new average. The formula for simple moving
average is expressed as:

Σ=(Most recent data values)


Number of time periods
Example
Mary Insurance Agency would like to get an accurate forecast of demand for life
insurance with face value in the amount of Php 2,000,000. Based on the sales records
from this sale agency in Las Pinas City, the following data has been accumulated
form the past 11 months. The owner of the agency wanted to look at the three- and
five-month moving average forecast.

Months Jan Feb Mar Apr May June July Aug Sept Oct Nov

Insurance 30 100 80 150 70 50 90 60 180 30 90


Exponential
04 Smoothing
Exponential Smoothing
Ideally, in making a forecast, we want to use the most current data and,
at the same time, use enough observations to obtain a smooth random
fluctuation. Exponential smoothing is a technique that is perfectly
designed to satisfy the two objectives. The simplest exponential
smoothing formula is as follows:

Where:
= the forecast for the next period
= actual demand in the present period
= the previously determined forecast for the present period
a = a weighting factor referred to as the smoothing constant
Example
Mary Insurance Agency would like to get an accurate forecast of demand for life
insurance with face value in the amount of Php 2,000,000. Based on the sales records
from this sale agency in Las Pinas City, the following data has been accumulated
form the past 11 months. The owner of the agency wanted to look at the three- and
five-month moving average forecast.

Months Jan Feb Mar Apr May June July Aug Sept Oct Nov

Insurance 30 100 80 150 70 50 90 60 180 30 90


Conclusion
Moving averages and exponential smoothing
are two of the most commomnly used models
in time-series analysis. The two models are
useful in smoothing historical data and are
also generally used to make a single
prediction one time period ahead
Thank you!
CRÉDITOS: Esta plantilla para presentaciones es una creación de Slidesgo, e incluye iconos de
Flaticon, infografías e imágenes de Freepik y contenido de José Antonio Cuenca

You might also like