A few weeks ago, I wrote about how and why I was learning Machine Learning, mainly through Andrew Ng's Coursera course. Machine Learning is built on prerequisites, so much so that learning by first principles seems overwhelming. Do you really need to spend a month learning linear algebra? You'll be okay if you have some math and programming experience. You really just have to be familiar with Sigma notation and be able to express it in a for loop. Sure, your assignments will take longer to complete and the first few times you see those giant equations your head will spin, but you can do this! Calculus is not even required.

Anomalies in healthcare claims data can be indicative of possible fraudulent activities, contributing to a significant portion of overall healthcare costs. Medicare is a large government run healthcare program that serves the needs of the elderly in the United States. The increasing elderly population and their reliance on the Medicare program create an environment with rising costs and increased risk of fraud. The detection of these potentially fraudulent activities can recover costs and lessen the overall impact of fraud on the Medicare program. In this paper, we propose a new method to detect fraud by discovering outliers, or anomalies, in payments made to Medicare providers. We employ a multivariate outlier detection method split into two parts. In the first part, we create a multivariate regression model and generate corresponding residuals. In the second part, these residuals are used as inputs into a generalizable univariate probability model. We create this Bayesian probability model using probabilistic programming. Our results indicate our model is robust and less dependent on underlying data distributions, versus Mahalanobis distance. Moreover, we are able to demonstrate successful anomaly detection, within Medicare specialties, providing meaningful results for further investigation.

There is intense interest in applying machine learning to problems of causal inference in fields such as healthcare, economics and education. In particular, individual-level causal inference has important applications such as precision medicine. We give a new theoretical analysis and family of algorithms for predicting individual treatment effect (ITE) from observational data, under the assumption known as strong ignorability. The algorithms learn a "balanced" representation such that the induced treated and control distributions look similar. We give a novel, simple and intuitive generalization-error bound showing that the expected ITE estimation error of a representation is bounded by a sum of the standard generalization-error of that representation and the distance between the treated and control distributions induced by the representation. We use Integral Probability Metrics to measure distances between distributions, deriving explicit bounds for the Wasserstein and Maximum Mean Discrepancy (MMD) distances. Experiments on real and simulated data show the new algorithms match or outperform the state-of-the-art.

Unlike evaluating the accuracy of models that predict a continuous or discrete dependent variable like Linear Regression models, evaluating the accuracy of a classification model could be more complex and time-consuming. Before measuring the accuracy of classification models, an analyst would first measure its robustness with the help of metrics such as AIC-BIC, AUC-ROC, AUC- PR, Kolmogorov-Smirnov chart, etc. The next logical step is to measure its accuracy. To understand the complexity behind measuring the accuracy, we need to know few basic concepts. E.g. – A classification model like Logistic Regression will output a probability number between 0 and 1 instead of the desired output of actual target variable like Yes/No, etc.

Thank you very much, as for your question, I meant that I have an univariate logistic regression model (i.e., with only one dependent binary variable), where the dependent variable must be explained by a number of binary independent variables (1,0). I have no problem when the independent variables are continuous in nature and normally distributed, because there is Hsieh (1998) who said that you can obtain the total sample size basing on the multiple correlation coefficient between Xi and the remaining predictors... However I didn't find anything like that for the model that I talked about above. So I hope to find in APPLIED LOGISTIC REGRESSION what I looking for.

Following my post about logistic regressions, Ryan got in touch about one bit of building logistic regressions models that I didn't cover in much detail – interpreting regression coefficients. This post will hopefully help Ryan (and others) out. I'd love to see more about interpreting the glm coefficients. Coefficients are what a line of best fit model produces. A line of best fit (aka regression) model usually consist of an intercept (where the line starts) and the gradients (or slope) for the line for one or more variables.

The link between different psychophysiological measures during emotion episodes is not well understood. To analyse the functional relationship between electroencephalography (EEG) and facial electromyography (EMG), we apply historical function-on-function regression models to EEG and EMG data that were simultaneously recorded from 24 participants while they were playing a computerised gambling task. Given the complexity of the data structure for this application, we extend simple functional historical models to models including random historical effects, factor-specific historical effects, and factor-specific random historical effects. Estimation is conducted by a component-wise gradient boosting algorithm, which scales well to large data sets and complex models.

No need to apologize for not using "proper" weights.

Hello, I am working on a highly imbalanced dataset (negative examples over 20K and positive examples about 100). I am trying to build a logistic regression model. My current approach includes undersampling of negative examples. However with this approach there are a couple of problems: 1) Several LR models are possible with different samples. How to generalize these models and interpret the output?

Regression is a time-tested manner for approximating relationships among a given collection of data, and the recipient of unhelpful naming via unfortunate circumstances. Linear regression is a simple algebraic tool which attempts to find the "best" (generally straight) line fitting 2 or more attributes, with one attribute (simple linear regression), or a combination of several (multiple linear regression), being used to predict another, the class attribute. A set of training instances is used to compute the linear model, with one attribute, or a set of attributes, being plotted against another. The model then attempts to identify where new instances would lie on the regression line, given a particular class attribute. It is often confusing for people without a sufficient math background to understand how matrix multiplication fits into linear regression.