Fact Based List:
Vincent Granville: The 8 Worst Predictive Modeling Techniques
Submitted by Anonymous on Mon, 10/01/2012 - 15:13
- Linear regression. Relies on normal, heteroscedasticity & other assumptions, doesn’t capture highly non-linear, chaotic patterns. Prone to over-fitting. Parameters difficult to interpret.
- Traditional decision trees. Very large decision trees are very unstable and impossible to interpret, and prone to over-fitting.
- Linear discriminant analysis. Used for supervised clustering. Bad technique because it assumes that clusters do not overlap, and are well separated by hyper-planes
- K-means clustering. Used for clustering, tends to produce circular clusters. Does not work well with data points that are not a mixture of Gaussian distributions.
- Neural networks. Difficult to interpret, unstable, subject to over-fitting.
- Maximum Likelihood estimation. Requires data to fit with a prespecified probabilistic distribution. Not data-driven. In many cases the pre-specified Gaussian distribution is terrible fit for data.
- Density estimation in high dimensions. Subject to what is referred to as the curse of dimensionality.
- Naive Bayes. Used e.g. in fraud and spam detection, and for scoring. Assumes that variables are independent, if not it will fail miserably.
Source: AnalyticBridge
Source URL: http://www.analyticbridge.com/profiles/blogs/the-8-worst-pre...
List Ratings: |
Lists You Might Also Be Interested In
- Partners HealthCare: “Disruptive Dozen” Technologies for 2015 at the World Medical Innovation Forum
- ECRI: Top 10 List of Health Technology Hazards for 2016
- Chip Howard: 3 Common Accountable Care Tech Challenges
- ECRI Institute's Top 10 Technology Issues for Hospital Leaders
- 11 Prominent Players in the Field of Wearable Technology in Healthcare
Login or register to post comments