1 / 18

# Defn : Forecasting is the art and science of predicting future events. - PowerPoint PPT Presentation

Defn : Forecasting is the art and science of predicting future events. 1. Identify the. 2. Collect. 3. Plot data and. purpose of forecast. historical data. identify patterns. 9. Adjust forecast based. 8a. Forecast over. on additional qualitative.

I am the owner, or an agent authorized to act on behalf of the owner, of the copyrighted work described.

## PowerPoint Slideshow about ' Defn : Forecasting is the art and science of predicting future events.' - desma

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.

- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript

Defn: Forecasting is the art and

science of predicting future

events.

2. Collect

3. Plot data and

purpose of forecast

historical data

identify patterns

8a. Forecast over

planning horizon

information and insight

Forecasting Process

5. Develop / compute forecast for

6. Check forecast accuracy

4. Select a forecast model that

period of historical data

with one or more measures

seems appropriate for data

8b. Select new forecast model or

7.

Is accuracy

of forecast

acceptable?

10. Monitor results and

measure forecast accuracy

2. Collect

3. Plot data and

purpose of forecast

historical data

identify patterns

8a. Forecast over

planning horizon

information and insight

Forecasting Process

5. Develop / compute forecast for

6. Check forecast accuracy

4. Select a forecast model that

period of historical data

with one or more measures

seems appropriate for data

8b. Select new forecast model or

7.

Is accuracy

of forecast

acceptable?

10. Monitor results and

measure forecast accuracy

• Time Series Models

• Associative Models

• Qualitative

Assumption: One or more variables can be identified which has a relationship with demand.

Approaches: Simple Linear Regression

Multiple Linear Regression

Defn: A time-ordered sequence of

observations that have been taken

at regular intervals.

Examples: past monthly demands,

past annual demands.

Assumption: Future values can be estimated from past values of the series.

• Naïve Approach

• Moving Averages

• Exponential Smoothing

• Trend Projection

2. Collect

3. Plot data and

purpose of forecast

historical data

identify patterns

8a. Forecast over

planning horizon

information and insight

Forecasting Process

5. Develop / compute forecast for

6. Check forecast accuracy

4. Select a forecast model that

period of historical data

with one or more measures

seems appropriate for data

8b. Select new forecast model or

7.

Is accuracy

of forecast

acceptable?

10. Monitor results and

measure forecast accuracy

• Trend

• gradual, long-term up or down movement

• Cycles

• up & down movement repeating over long time frame

• Seasonality

• periodic oscillation in demand which repeats based on calendar schedule (days, weeks, months or quarters)

• Random movements- follow no pattern

• Class Exercise:Decompose the TimeSeriesGraphs.xls time series into the behavioral components

• Upward trend

• Downward trend

• Seasonality

• Random Variation

• Upward trend

• Downward trend

• Seasonality

• Random Variation

• Upward trend

• Downward trend

• Seasonality

• Random Variation

• Three possible models to choose from when there is no seasonality and not a strong trend pattern:

• Naïve Approach

• n Period Moving Average

• Exponential Smoothing

Demand in

Previous

n

Periods

MA

n

nperiod Moving Average Method

• MA is a series of arithmetic means

• Used if little or no trend

• Used oftenfor smoothing

2. Collect

3. Plot data and

purpose of forecast

historical data

identify patterns

8a. Forecast over

planning horizon

information and insight

Forecasting Process

5. Develop / compute forecast for

6. Check forecast accuracy

4. Select a forecast model that

seems appropriate for data

period of historical data

with one or more measures

8b. Select new forecast model or

7.

Is accuracy

of forecast

acceptable?

10. Monitor results and

measure forecast accuracy

• k= # of historical periods forecasted

• Mean Absolute Percent Deviation (MAPD)

• New Forecast

• = Last Period Forecast + Correction for error made last period

• = Last Period Forecast + α (Last Period Demand – Last Period Forecast)

• Class exercise:

• Identify the best α for Smooth.xls time series

• You want to achieve:

• No pattern or direction in forecast error

• Error = (Ai - Fi) = (Actual - Forecast)

• Seen in plots of errors over time

• Smallest forecast error