Scribd
Upload
21 views6 pages

Forecasting

Forecasting

Uploaded by

tusharshk7
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
21 views6 pages

Forecasting

Forecasting

Uploaded by

tusharshk7
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 6

Forecasting

There are many types of forecasting models. They differ in their degree of
complexity, the amount of data they use, and the way they generate the forecast.
However, some features are common to all forecasting models. They include the
following:

● Forecasts are rarely perfect: Forecasting the future involves uncertainty.


Therefore, it is almost impossible to make a perfect prediction. The goal of
forecasting is to generate good forecasts on the average over time and to
keep forecast errors as low as possible.
● Every forecast should include an estimate of error: Since forecasts are
expected to be wrong, the real question is “By how much?” Every forecast
should include an estimate of error often expressed as a percentage (plus
and minus) of the forecast or as a range between maximum and minimum
values.
● Forecasts are more accurate for groups or families of items rather than
for individual items: When items are grouped together, their individual
high and low values can cancel each other out. The data for a group of
items can be stable even when individual items in the group are very
unstable. Consequently, one can obtain a higher degree of accuracy when
forecasting for a group of items rather than for individual items. For
example, you cannot expect the same degree of accuracy if you are
forecasting sales of long-sleeved hunter green polo shirts that you can
expect when forecasting sales of all polo shirts.
● Forecasts are more accurate for shorter than longer time horizons: The
shorter the time horizon of the forecast, the lower the degree of
uncertainty. Data do not change very much in the short run. As the time
horizon increases, however, there is a much greater likelihood that
changes in established patterns and relationships will occur. For example,
it is much harder to predict sales of a product two years from now than to
predict sales two weeks from now.

There are three types of forecasts:

1. Judgmental: Uses subjective inputs like executive opinion, sales force


opinions, consumer surveys, Delphi method, etc.
2. Time Series: Uses historical data assuming future value is dependent on
the past. It includes methods like naïve forecast, moving averages,
weighted moving average, exponential smoothing, Holt’s method,
Winters’s method etc.
3. Associative Forecasting: Uses explanatory variables to predict the future.
It includes methods like predictor variable, regression, least squares line

Operations Book
Forecasting Techniques
There are many forecasting methods, but they can usually be classified into:

Qualitative Techniques
Qualitative forecasting methods, often called judgmental methods, are methods
in which the forecast is made subjectively by the forecaster. They are educated
guesses by forecasters or experts based on intuition, knowledge, and
experience. When you decide, based on your intuition, that a particular team is
going to win a cricket game, you are making a qualitative forecast. Because
qualitative methods are made by people, they are often biased.

Quantitative Techniques
Quantitative forecasting methods, on the other hand, are based on mathematical
modelling. Because they are mathematical, these methods are consistent. The
same model will generate the exact same forecast from the same set of data
every time.

These methods are also objective. They do not suffer from the biases found in
qualitative forecasting. Finally, these methods can consider a lot of information at
one time. Because people have limited information-processing abilities and can
easily experience information overload, they cannot compete with
mathematically generated forecasts in this area.

Types of Qualitative Techniques

Operations Book
Types of Quantitative Techniques
• Moving Average

Procedure is to calculate the average company sales for previous years. Moving
averages name is due to dropping sales in the oldest period and replacing it by
sales in the newest period.

Example: 2-period moving average

• Weighted Moving Average

The weighted moving average is calculated by multiplying each datum in your


series by a different ratio and then taking the sum of those products. Weighted
averages assign a heavier weighting to more current data points since they are
more relevant than data points in the distant past. The sum of the weighting
should add up to 1 (or 100%).

Example:

Operations Book
The weighted average is calculated by multiplying the given price by its
associated weighting and then summing the values. In the example above, the
weighted 5-day moving average would be $90.62.

Calculation
((90.9*(5/15)) + (90.36*(4/15)) + (90.28*(3/15)) + (90.83*(2/15)) + (90.91*(1/15)))

Exponential smoothing

It is not necessary to keep months of history to get a moving average because


the previously calculated forecast has already allowed for this history. Therefore,
the forecast can be based on the old calculated forecast and the new data.

New forecast = (alpha) (latest demand) + (1 – alpha) (previous forecast)

Alpha is known as smoothing constant. When more weight is to be assigned to


latest demand, alpha is more than 0.5 and vice versa.
Exponential smoothing provides a routine method for regularly updating item
forecasts. It works quite well when dealing with stable items. Generally, it has
been found satisfactory for short- range forecasting. It is not satisfactory where
the demand is low or intermittent. Exponential smoothing will detect trends,
although the forecast will lag actual demand if a definite trend exists.

If a trend exists, it is possible to use a slightly more complex formula called


double exponential smoothing. This technique uses the same principles but
notes whether each successive value of the forecast is moving up or down on a
trend line.

Level, Trend, Seasonality and Noise

A useful abstraction for selecting forecasting methods is to break a time series


down into systematic and unsystematic components.

Systematic: Components of the time series that have consistency or recurrence


and can be described and modeled.

Non-Systematic: Components of the time series that cannot be directly


modeled.

A given time series is thought to consist of three systematic components


including level, trend, seasonality, and one non- systematic component called
noise.
These components are defined as follows:

Level: The average value in the series.

Trend: The increasing or decreasing value in the series.

Seasonality: The repeating short-term cycle in the series.

Operations Book
Noise: The random variation in the series

These components can be combined to form the time series either additively or
multiplicatively, i.e:

Additive: y(t) = Level + Trend + Seasonality + Noise Multiplicative: y(t) = Level *


Trend * Seasonality * Noise

Comparison of different Time Series Models

Quantitative v/s Qualitative Techniques

Operations Book
Forecast Error
In statistics, a forecast error is the difference between the actual or real and the
predicted or forecast value of a time series or any other phenomenon of interest.
Since the forecast error is derived from the same scale of data, comparisons
between the forecast errors of different series can only be made when the series
are on the same scale.

Forecast error is the difference between actual demand and forecast demand.
Various techniques to map the efficiency of model or find accuracy of model:

• Mean absolute deviation (MAD)

It takes the absolute value of forecast errors and averages them over the entirety
of the forecast time periods. Taking an absolute value of a number disregards
whether the number is negative or positive and, in this case, avoids the positives
and negatives cancelling each other out.

• Mean absolute percentage error (MAPE)

It is the most common measure of forecast error. MAPE functions best when
there are no extremes to the data (including zeros). MAPE is the average absolute
percent error for each time period or forecast minus actuals divided by actuals.

• Mean Squared Error (MSE)

Average of squared error i.e., ratio of sum of squares of all errors and total no.
of data points -1.

Operations Book

You might also like