• Skip to main content
  • Skip to secondary menu
  • Skip to primary sidebar
  • Skip to footer
  • Articles
  • News
  • Events
  • Advertize
  • Jobs
  • Courses
  • Contact
  • (0)
  • LoginRegister
    • Facebook
    • LinkedIn
    • RSS
      Articles
      News
      Events
      Job Posts
    • Twitter
Datafloq

Datafloq

Data and Technology Insights

  • Categories
    • Big Data
    • Blockchain
    • Cloud
    • Internet Of Things
    • Metaverse
    • Robotics
    • Cybersecurity
    • Startups
    • Strategy
    • Technical
  • Big Data
  • Blockchain
  • Cloud
  • Metaverse
  • Internet Of Things
  • Robotics
  • Cybersecurity
  • Startups
  • Strategy
  • Technical

5 Painful Prediction Pitfalls

Lars Trieloff / 3 min read.
March 17, 2015
Datafloq AI Score
×

Datafloq AI Score: 74

Datafloq enables anyone to contribute articles, but we value high-quality content. This means that we do not accept SEO link building content, spammy articles, clickbait, articles written by bots and especially not misinformation. Therefore, we have developed an AI, built using multiple built open-source and proprietary tools to instantly define whether an article is written by a human or a bot and determine the level of bias, objectivity, whether it is fact-based or not, sentiment and overall quality.

Articles published on Datafloq need to have a minimum AI score of 60% and we provide this graph to give more detailed information on how we rate this article. Please note that this is a work in progress and if you have any suggestions, feel free to contact us.

floq.to/9hdhX

Through my work with customers and data scientists building predictive applications based on the Blue Yonder Platform and integrating our standard products, such as Forward Demand, into customer workflows, Ive encountered a few common prediction pitfalls. While these might not ruin the scientific validity of the model, they may affect the practicality of and the applicability to the business case.

Predictive applications require applied data science and cannot live outside of their application domain. Watching out for these common prediction pitfalls will help you deliver predictive applications that are relevant to the business and help your customers make (or save) money:


Interested in what the future will bring? Download our 2023 Technology Trends eBook for free.

Consent

  1. Forecast what you can control. In a predictive application, forecasts serve to make automated or semi-automated decisions. This means that for the best results, align the decision horizon and granularity as well as the prediction horizon and granularity as much as possible. For instance, if a predictive application, such as Forward Demand, is controlling a replenishment workflow for slow-moving goods in which deliveries arrive on a weekly basis, making daily demand forecasts will not create any benefit to the business and might actually reduce value by reducing statistical significance of the model and shortening the prediction horizon. If your deliveries are occurring on a weekly basis then you need weekly predictions.
  2. Beware of false aggregations. Powerful predictive applications, such as Forward Demand, not only offer simple forecasts but they can also predict the can also predict the individual probability of each single event, taking into account a wide range of influencing factors. In fact, all predictive applications based on NeuroBayes come with this capability. A good example is the prediction for failure of a machine part vs. failure of the entire machine. A dual-engine machine has an estimated 95th-percentile failure date of May 1st for one engine and May 30th for the second engine. Correctly predicting the expected failure date of the machine is not a simple aggregation (May 15th) but either requires knowledge of the inner workings of the machine or a prediction model that handles the entire machine.
  3. The absence of proof is not the proof of absence. Consider a demand forecasting scenario: when analyzing individual point of purchase data, no purchases can be seen in the second half of the week. Treating the absence of purchases as proof for the absence of demand could lead to costly mistakes. A different explanation might be that stores are running out of stock in the middle of the week, so that any demand in the second half of the week goes unsatisfied. In this case, treating the absence of proof as the proof of absence will lead to severe out-of-stock situations with huge opportunity costs.
  4. Consider the cost of uncertainty. Speaking of cost and opportunity costs, many predictive applications must make decisions in the light of hard tradeoffs and with great uncertainty. Lets assume we are scoring the fraud risk for an online customer. Providing the customer with the ability to pay by invoice sets us up for the risk of losing the entire value of the merchandise purchased. On the other hand, denying the customer the ability to pay by invoice might lead to the customer abandoning the shopping cart. Thereby losing us the opportunity to win a customer with considerable customer lifetime value. In order to make decisions like these on a regular, automated basis, multiple predictions need to be combined: fraud risk, shopping cart abandonment risk and customer lifetime value. Leaving these factors and uncertainties unaccounted for can lead to a gross misrepresentation of risk, typically resulting in retailers’ loss of revenue as fraud risks are overestimated.
  5. There are many wrong ways to calculate forecast quality. As soon as a predictive model is in place, the question of accuracy of the model comes up. Unfortunately, many intuitive approaches to assessing forecast quality can lead business customers astray. Simply comparing the predictions for a few top sellers can lead to both an overestimation of forecast errors (look, the model has been off by more than 1.000 units) while underestimating the overall accuracy across the entire portfolio (only 2% deviation). The topic of selecting and designing the right measure of quality is one of the key issues of consideration when building predictive applications. So much so, that my colleagues Michael Feindt and Ulrich Kerzel actually wrote a (German) book to help data scientists and business users navigate the maze of forecast quality metrics.

Building and deploying predictive applications has an incredible and transformative impact on businesses and helps them accelerate their transition to becoming a data-driven enterprise, but it is not without challenges. Steering clear of these five pitfalls will help you to get up and running faster and deliver better results.

More about predictive applications and their use can be found here. 

Categories: Big Data
Tags: Big Data, predictive analytics

About Lars Trieloff

Before Lars Trieloff joined Blue Yonder he has been responsible for product management of Marketing Cloud at Adobe. Prior to that at Day Software, Lars was responsible for product management of Day's digital asset management and social community products, which expanded the breadth of the shared content platform. He holds a MSc of Software Systems Engineering from Hasso Plattner Institute in Potsdam. Lars is married and the father of four cildren.

Primary Sidebar

E-mail Newsletter

Sign up to receive email updates daily and to hear what's going on with us!

Publish
AN Article
Submit
a press release
List
AN Event
Create
A Job Post

Related Articles

How to Validate OpenAI GPT Model Performance with Text Summarization (Part 1)

March 29, 2023 By mark

Big Data & AI World, Singapore

March 29, 2023 By r.chan

Velocity Data and Analytics Summit, UAE

March 29, 2023 By shiwangi-7725

Related Jobs

  • Software Engineer | South Yorkshire, GB - February 07, 2023
  • Software Engineer with C# .net Investment House | London, GB - February 07, 2023
  • Senior Java Developer | London, GB - February 07, 2023
  • Software Engineer – Growing Digital Media Company | London, GB - February 07, 2023
  • LBG Returners – Senior Data Analyst | Chester Moor, GB - February 07, 2023
More Jobs

Tags

AI Amazon analysis analytics app application Artificial Intelligence BI Big Data blockchain business China Cloud Companies company costs crypto customers Data development digital environment experience finance future Google+ government information learning machine learning market mobile Musk news public research security share social social media software startup strategy technology twitter

Related Events

  • 6th Middle East Banking AI & Analytics Summit 2023 | Riyadh, Saudi Arabia - May 10, 2023
  • Data Science Salon NYC: AI & Machine Learning in Finance & Technology | The Theater Center - December 7, 2022
  • Big Data LDN 2023 | Olympia London - September 20, 2023
More events

Related Online Courses

  • Big Data & AI World, Singapore
  • Velocity Data and Analytics Summit, UAE
  • Intel AI Fundamentals
More courses

Footer


Datafloq is the one-stop source for big data, blockchain and artificial intelligence. We offer information, insights and opportunities to drive innovation with emerging technologies.

  • Facebook
  • LinkedIn
  • RSS
  • Twitter

Recent

  • How to Validate OpenAI GPT Model Performance with Text Summarization (Part 1)
  • What is Enterprise Application Integration (EAI), and How Should Your Company Approach It?
  • 5 Best Data Engineering Projects & Ideas for Beginners
  • Personalization Vs. Hyper-Personalization: Benefits, Limitations and Potential
  • Explaining data products lifecycle and their scope in management

Search

Tags

AI Amazon analysis analytics app application Artificial Intelligence BI Big Data blockchain business China Cloud Companies company costs crypto customers Data development digital environment experience finance future Google+ government information learning machine learning market mobile Musk news public research security share social social media software startup strategy technology twitter

Copyright © 2023 Datafloq
HTML Sitemap| Privacy| Terms| Cookies

  • Facebook
  • Twitter
  • LinkedIn
  • WhatsApp

In order to optimize the website and to continuously improve Datafloq, we use cookies. For more information click here.

settings

Dear visitor,
Thank you for visiting Datafloq. If you find our content interesting, please subscribe to our weekly newsletter:

Did you know that you can publish job posts for free on Datafloq? You can start immediately and find the best candidates for free! Click here to get started.

Not Now Subscribe

Thanks for visiting Datafloq
If you enjoyed our content on emerging technologies, why not subscribe to our weekly newsletter to receive the latest news straight into your mailbox?

Subscribe

No thanks

Privacy Overview

This website uses cookies so that we can provide you with the best user experience possible. Cookie information is stored in your browser and performs functions such as recognising you when you return to our website and helping our team to understand which sections of the website you find most interesting and useful.

Necessary Cookies

Strictly Necessary Cookie should be enabled at all times so that we can save your preferences for cookie settings.

If you disable this cookie, we will not be able to save your preferences. This means that every time you visit this website you will need to enable or disable cookies again.

Marketing cookies

This website uses Google Analytics to collect anonymous information such as the number of visitors to the site, and the most popular pages.

Keeping this cookie enabled helps us to improve our website.

Please enable Strictly Necessary Cookies first so that we can save your preferences!