Loading presentation...

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

Make your likes visible on Facebook?

Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.

No, thanks

forecasting

No description
by

thomas cesena

on 13 September 2012

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of forecasting

MKTG 188 - Cesena, Thomas - Row 4 Forecasting Methods Forecasting: A blend of art and science Use observed results from previous period as forecast for the upcoming period 8 - naive approach 6 steps of forecasting process :
Determine the purpose
Establish a time horizon
Select technique
Gather and analyze data
Prepare the forecast
Monitor Elements of a good forecast:
Timely
Accurate
Reliable
Measured in meaningful units
In writing
Easy to understand Based on Judgements and Opinions
Originated in 1948 with Rand Corporation 9 - Delphi method Methodology: Series of surveys given to managers and staff, each formulated from the previous.

*Each member must have ability to contribute meaningfully*

Goal: Achieve a consensus forecast http://trendsoutheast.org/2011/wp-content/uploads/2011/12/2513262775_672082cb0c.jpg
http://highered.mcgraw-hill.com/sites/dl/free/0072443901/21708/ste43901_ch03qxd.pdf
http://www.ops.fhwa.dot.gov/publications/fhwahop12009/sec5.htm http://highered.mcgraw-hill.com/sites/dl/free/0072443901/21708/ste43901_ch03qxd.pdf Averages actual recent values and updates as new values are available 7 - Simple Moving Average Advantages:
Little to no cost
Quick
Easily understood
Best in short run Disadvantages:
Not highly accurate
Has lag of one period
Doesn't smooth data
Not as reliable in long run Advantages:
Easy to compute and understand
Reflects recent change
Safe because forecasts stay within historical data Disadvantage:
All values carry the same weight
Does not emphasize most recent events Similar to simple moving average except more recent values in a data series are given more weight(importance) 6 - Weighted moving average Can incorporate as many data points as desired
More data points = less responsive
Less data points = more responsive Advantages:
More reflective of recent occurrences than a simple moving average
More likely to pick up trends Weighted averaging method based on previous forecast plus a percentage of the forecast error 5 - Exponential smoothing Provides a look at the linear relationship between an independent variable and dependent variable. 2 - Simple linear regression Advantages:
Requires limited amount of data
Relatively simple
Expandable to trend and seasonal models Disadvantages:
Lags behind actual data
Incapable of including causal factors Advantages:
Allows anonymity
Less biased
Flexible and applicable variety of issues Disadvantages:
May not be representative
Can be time consuming Formula: Y = a + bX + e
Y = Independent variable
a = Intercept of line
b = Slope of the line
X = Dependent variable
e = Residual or error Advantages:
Can prove significant correlation between variables
Easy to use with computer software Disadvantages:
Possibility of outliers can throw off data Technique used to study the relationship of one dependent variable and multiple independent variables 1 - Multiple regression Advantages:
Allows explicit control
Can run multiple tests with computer software
Flexible (independent variables may be categorical or numerical) http://www.ehow.com/info_12070171_advantages-disadvantages-multiple-regression-model.html
http://arthritis-research.com/content/figures/ar1164-1-l.jpg Disadvantages:
Possibility of outliers
Multicollinearity http://labs.fme.aegean.gr/decision/images/stories/docs/healthcare_forecast.pdf http://highered.mcgraw-hill.com/sites/dl/free/0072443901/21708/ste43901_ch03qxd.pdf Example: Can be used in medical field to find correlations between different variables that may cause disease. Disadvantages:
Weights are chosen arbitrarily
Disregards historical data that could be relevant
May take some trial and error
Can produce false signals of trends http://labs.fme.aegean.gr/decision/images/stories/docs/healthcare_forecast.pdf Weights can range from 0 to 1
*The larger the weight, the more important (subjectively)
* Weights must add up to 1 http://highered.mcgraw-hill.com/sites/dl/free/0072443901/21708/ste43901_ch03qxd.pdf
http://highered.mcgraw-hill.com/sites/dl/free/0070951640/354829/lind51640_ch16.pdf Short term regular variations ranging occurring throughout the day, week, month, or year 3 - Seasonal Variation Examples:
Back to school supplies
Sporting goods for correlating seasons
Toy sales rise during Christmas season Advantages:
Easily identifiable with sufficient data
Firms can focus on marketing during these times
Firms can plan production levels accordingly Disadvantages:
Takes time to collect sufficient data
May disregard other factors Can be used for:
Stable series
Seasonal variations
Trend Simple moving average equation Alpha is closer to 0 - less responsive/smoother
Alpha is closer to 1 - more responsive/less smooth Exponential smoothing equation http://highered.mcgraw-hill.com/sites/dl/free/0070951640/354829/lind51640_ch16.pdf Occurs when there is a steady long term increase or decrease in the data 4 - Linear Trend Examples:
Trend in women's participation in the labor force Advantages:
Can be used to predict further than one period into the future
Acts as a "best fit" model
Simple to use Disadvantages:
Cannot always be used for long term predictions
Doesn't account for seasonal or cyclical trends Great method when there is no historical data to rely on 10 - Forecasts based on judgement and opinion Executive Opinions
Useful in long term planning and product development
Potential for great collaboration
*Group can fall victim to groupthink* Consumer Surveys
Allow a sample of consumer opinions
Firms receive feedback
*Costly and time consuming* Salesforce Opinions
Have direct contact with customers
Get an inside look at consumer's future plans
*May have a conflict of interest* http://highered.mcgraw-hill.com/sites/dl/free/0072443901/21708/ste43901_ch03qxd.pdf
http://www.docstoc.com/docs/3568096/Forecasting-using-trend-analysis
http://www.ssc.wisc.edu/~bhansen/390/390Lecture6.pdf Proposed Forecasting Method: Seasonality White flesh nectarine case 5 step procedure:
Determine demand for white flesh nectarines on east coast last season
Compare demand from last season to potential harvest this season
Determine if there will be a surplus or shortage
Contact east coast buyers with news
Allocate the nectarines as necessary
Example: Forecasting number of doctor visits for the following year Goal: produce forecast with lowest mean square error(MSE) "Best Fit" trend line Actual Data Weighted moving average Vs. Simple moving average
Full transcript