A Data Science Central Community

You may wonder how those mighty people do the fancy projection in their business plan, there are broadly three methods of forecast and it is suitable in different situations: 1. Bottom up: 2. Top down guestimate, and 3. Top down forecast.

1. Bottom up

As its name suggest, forecast is down on name by name basis by the sales person, it is more often to have bottom-up forcast when the client numbers are smaller but High value Eg investment banking often uses this approach in its sales forcast. However it will be highly subject to the gaming of the provider- they would put a lower number in the estimate and you find no basis to challenge it as there is only one person who knows the truth. One famous bottom-up indicator is purchasing management indicator.

2. Top down guesstimate

By the gut feeling of an authority or simply done by screaming. Someone with the authority from the organization apply some random growth rate as the objective, then the rest of the people work backward for the solutions. I have done similar exercises like this before, the challenges are to allocate back how much are needed from each of the constitute. Sometimes goal seeking are needed to guess back the combination needed.

3. Top down forecast

By linking the management data and relevant leading indicators, we can build the model that describes the causation of the factors ( could be economic indicators) vs the internal management data (could be sales volume of products). How to pick the relevant factors for forecasting? For example, if you want to forecast the GDP of a country, you can start by looking at the major industries within the economy. The method of decomposition by finding the relevant indicators are a useful rule of thumb. Ok then you get the model, the next problem comes: How to translate it into common languages where the CEO, CMO understand? The challenges are 1. you do not know how to explain the coefficient and 2. the model structure can be unstable in the turmoil.

Problem one is the biggest challenge I had in my early career. This damper the ability to use model for forecast. My lesson learnt are: 1. You have to meet the right boss to use model; 2. You have to ride on the right wave (following the management's strategic direction when applying the model, not to stick to what the model can do). 3. Study hard: You need to spend a huge amount of time to understand the model and explain in details and steps by steps how it works. 4. Create confidence by modeling your company's common wisdom into model. e.g. if most people usually think that there should be cross selling between two products, model it using predictive modeling and prove it.

To exemplify the instability of model, one typical example is that the GDP of china vs government spending. Being the major component of GDP, using government spending to estimate GDP means you assume that government spending will always be used as the main tool to expand/ sustain the economy- keeping GDP growing. It is true during the financial tsunami and lasts for only a two to three years. The point here is that once you start using the model, there should be routine to validate it on a regular basis.

Should there be the forth method? Random numbers generation- forecast may have not use to some people.

© 2021 TechTarget, Inc. Powered by

Badges | Report an Issue | Privacy Policy | Terms of Service

**Most Popular Content on DSC**

To not miss this type of content in the future, subscribe to our newsletter.

- Book: Applied Stochastic Processes
- Long-range Correlations in Time Series: Modeling, Testing, Case Study
- How to Automatically Determine the Number of Clusters in your Data
- New Machine Learning Cheat Sheet | Old one
- Confidence Intervals Without Pain - With Resampling
- Advanced Machine Learning with Basic Excel
- New Perspectives on Statistical Distributions and Deep Learning
- Fascinating New Results in the Theory of Randomness
- Fast Combinatorial Feature Selection

**Other popular resources**

- Comprehensive Repository of Data Science and ML Resources
- Statistical Concepts Explained in Simple English
- Machine Learning Concepts Explained in One Picture
- 100 Data Science Interview Questions and Answers
- Cheat Sheets | Curated Articles | Search | Jobs | Courses
- Post a Blog | Forum Questions | Books | Salaries | News

**Archives:** 2008-2014 |
2015-2016 |
2017-2019 |
Book 1 |
Book 2 |
More

**Most popular articles**

- Free Book and Resources for DSC Members
- New Perspectives on Statistical Distributions and Deep Learning
- Time series, Growth Modeling and Data Science Wizardy
- Statistical Concepts Explained in Simple English
- Machine Learning Concepts Explained in One Picture
- Comprehensive Repository of Data Science and ML Resources
- Advanced Machine Learning with Basic Excel
- Difference between ML, Data Science, AI, Deep Learning, and Statistics
- Selected Business Analytics, Data Science and ML articles
- How to Automatically Determine the Number of Clusters in your Data
- Fascinating New Results in the Theory of Randomness
- Hire a Data Scientist | Search DSC | Find a Job
- Post a Blog | Forum Questions

## You need to be a member of AnalyticBridge to add comments!

Join AnalyticBridge