Typically, these bad techniques are still widely used.
- Linear regression. Relies on the normal, heteroscedasticity and other assumptions, does not capture highly non-linear, chaotic patterns. Prone to over-fitting. Parameters difficult to interpret. Very unstable when independent variables are highly correlated. Fixes: variable reduction, apply a transformation to your variables, use constrained regression (e.g. ridge or Lasso regression)
- Traditional decision trees. Very large decision trees are very unstable and impossible to interpret, and prone to over-fitting. Fix: combine multiple small decision trees together instead of using a large decision tree.
- Linear discriminant analysis. Used for supervised clustering. Bad technique because it assumes that clusters do not overlap, and are well separated by hyper-planes. In practice, they never do. Use density estimation techniques instead.
- 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 your data to fit with a prespecified probabilistic distribution. Not data-driven. In many cases the pre-specified Gaussian distribution is a terrible fit for your data.
- Density estimation in high dimensions. Subject to what is referred to as the curse of dimensionality. Fix: use (non parametric) kernel density estimators with adaptive bandwidths.
- Naive Bayes. Used e.g. in fraud and spam detection, and for scoring. Assumes that variables are independent, if not it will fail miserably. In the context of fraud or spam detection, variables (sometimes called rules) are highly correlated. Fix: group variables into independent clusters of variables (in each cluster, variables are highly correlated). Apply naive Bayes to the clusters. Or use data reduction techniques. Bad text mining techniques (e.g. basic «word» rules in spam detection) combined with naive Bayes produces absolutely terrible results with many false positives and false negatives.
And remember to use sound cross-validations techniques when testing models!
The reasons why such poor models are still widely used are:
- Many University curricula still use outdated textbooks, thus many students are not exposed to better data science techniques
- People using black-box statistical software, not knowing the limitations, drawbacks, or how to correctly fine-tune the parameters and optimize the various knobs, or not understanding what the software actually produces.
- Government forcing regulated industries (pharmaceutical, banking, Basel) to use the same 30-year old SAS procedures for statistical compliance. For instance, better scoring methods for credit scoring, even if available in SAS, are not allowed and arbitrarily rejected by authorities. The same goes with clinical trials analyses submitted to the FDA, SAS being the mandatory software to be used for compliance, allowing the FDA to replicate analyses and results from pharmaceutical companies.
- Modern data sets are considerably more complex and different than the old data sets used when these techniques were initially developed. In short, these techniques have not been developed for modern data sets.
- There’s no perfect statistical technique that would apply to all data sets, but there are many poor techniques.
In addition, poor cross-validation allows bad models to make the cut, by over-estimating the true lift to be expected in future data, the true accuracy or the true ROI outside the training set. Good cross validations consist in
- splitting your training set into multiple subsets (test and control subsets),
- include different types of clients and more recent data in the control sets (than in your test sets)
- check quality of forecasted values on control sets
- compute confidence intervals for individual errors (error defined e.g. as |true value minus forecasted value|) to make sure that error is small enough AND not too volatile (it has small variance across all control sets)