Talk:Predictive modelling

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Statistics (Rated Start-class, Mid-importance)
WikiProject icon

This article is within the scope of the WikiProject Statistics, a collaborative effort to improve the coverage of statistics on Wikipedia. If you would like to participate, please visit the project page or join the discussion.

Start-Class article Start  This article has been rated as Start-Class on the quality scale.
 Mid  This article has been rated as Mid-importance on the importance scale.
 

Needs major work[edit]

All of these sections are empty, or contain only a link to the main article. Each of them needs at least a brief write up, or else the headings need to be removed and the links moved elsewhere in the article.

  1. 1.1 Classification trees
  2. 1.2 Naive Bayes
  3. 1.3 k-nearest neighbor algorithm
  4. 1.4 Majority classifier
  5. 1.5 Support vector

FICO Score Accuracy section[edit]

I am removing the section(s) on FICO score accuracy. The section is not cited. I am not an expert on FICO, but I suspect that if it's supposed to indicate a certain percentage of defaults on average, it may mean that the FICO default rate will naturally rise in adverse economic environments, such as in 2009. If the section can be supported by a robust analysis of FICO scores, then it should be re-included. — Preceding unsigned comment added by 134.215.213.254 (talk) 19:31, 28 August 2013 (UTC)

Predictive analytics vs. predictive modelling[edit]

There is a separate wikipedia entry for Predictive analytics that appears to me to be covering largely the same material as this Predictive modeling entry. And it is odd that the predictive modelling entry doesn't contain any links to predictive analytics. To me, it looks like these two wikipedia entries should be combined. My idea would be to incorporate some of the material from the predictive modelling entry into the predictive analytics entry, then establish a "re-direct", so that when someone types in "predictive modeling" they are brought to predictive analytics. However, I'm an inexperienced wikipedia contributor, and am unaware of the general wikipedia conventions for consolidating articles like this. I also do not want to offend the authors who have put lots of hard work into the two entries. -- So I merely want to bring up the idea here for discussion to see what others think of the idea. Karl (talk) 15:33, 1 March 2014 (UTC)

The introduction to the entry for predictive modelling ends with the reference: "predictive modelling is synonymous with, or largely overlapping with, the field of machine learning, as it is more commonly referred to in academic or research and development contexts. When deployed commercially, predictive modelling is often referred to as predictive analytics." However, the entry for predictive analytics suggests that it is broader than predictive modelling and doesn't exclude non-commercial uses. I interpret the entry for machine learning to define it as a field of study, which uses predictive techniques. (I'm unfamiliar with any of these terms, but I'm inclined to agree that the entries for predictive modelling and predictive analytics should be merged.) rootsmusic (talk) 18:18, 11 August 2016 (UTC)

Dr. Gogas's comment on this article[edit]

Dr. Gogas has reviewed this Wikipedia page, and provided us with the following comments to improve its quality:


The article provides a rough introduction to predictive modelling, but in my opinion trying to mention to all possible forecasting methodologies is not productive. I would consider a more general description of the field and focus on the applied examples.


We hope Wikipedians on this talk page can take advantage of these comments and improve the quality of the article accordingly.

Dr. Gogas has published scholarly research which seems to be relevant to this Wikipedia article:


  • Reference : Gogas, Periklis & Papadimitriou, Theophilos & Plakandaras, Vasilios, 2013. "Forecasting the insolvency of U.S. banks using Support Vector Machines (SVM) based on Local Learning Feature Selection," DUTH Research Papers in Economics 2-2013, Democritus University of Thrace, Department of Economics.

ExpertIdeas (talk) 02:30, 4 September 2015 (UTC)

Non-neutral point of view in applications[edit]

The "notable failures" and "fundamental limitations" sections under "Applications" are written from the exclusive perspective of the 2008 financial crisis and the role that predictive modeling played. It sounds like the use of predictive modeling is being actively discouraged in light of these events. There is also a fair use of loaded language: "relying blindly on models that are essentially backforward looking in nature" (backfoward isn't proper English); "almost always wrong when the system involves people"; and the use of the word "spectacular" 3 times ("spectacular failures", "failed spectacularly", "spectacular debacle"). Looking through the article history, these sections appear to have been authored all by one person (https://en.wikipedia.org/wiki/Special:Contributions/Alpha1398).
I think this section should be rewritten to conform to the standards for neutral point of view. 171.66.68.154 (talk) 00:04, 1 April 2016 (UTC)Neil