<H1> Yet Another Blog in Statistical Computing </H1> |
<H1> Categories </H1> |
<H1> Top Posts & Pages </H1> |
<H1> R </H1> |
<H1> YAP: Yet Another Probabilistic Neural Network </H1> |
<H1> Improve General Regression Neural Network by Monotonic Binning </H1> |
<H1> GRNN with Small Samples </H1> |
<H1> GRNN vs. GAM </H1> |
<H1> Permutation Feature Importance (PFI) of GRNN </H1> |
<H1> Partial Dependence Plot (PDP) of GRNN </H1> |
<H1> Merge MLP And CNN in Keras </H1> |
<H1> Assess Variable Importance In GRNN </H1> |
<H1> Hyper-Parameter Optimization of General Regression Neural Networks </H1> |
<H1> Modeling Practices of Operational Losses in CCAR </H1> |
<H1> Develop Performance Benchmark with GRNN </H1> |
<H1> Dummy Is As Dummy Does </H1> |
<H1> Improve GRNN Efficiency by Weighting </H1> |
<H1> Yet Another R Package for General Regression Neural Network </H1> |
<H1> Monotonic Binning Driven by Decision Tree </H1> |
<H1> Chunk Averaging of GLM </H1> |
<H1> Latin Hypercube Sampling in Hyper-Parameter Optimization </H1> |
<H1> Parallel R: Socket or Fork </H1> |
<H1> WoE Transformation for Loss Given Default Models </H1> |
<H1> Faster Way to Slice Dataframe by Row </H1> |
<H1> Granular Weighted Binning by Generalized Boosted Model </H1> |
<H1> Why Use Weight of Evidence? </H1> |
<H1> More General Weighted Binning </H1> |
<H1> Binning with Weights </H1> |
<H1> Batch Deployment of WoE Transformations </H1> |
<H1> Batch Processing of Monotonic Binning </H1> |
<H1> Monotonic Binning with GBM </H1> |
<H1> Deployment of Binning Outcomes in Production </H1> |
<H1> A Summary of My Home-Brew Binning Algorithms for Scorecard Development </H1> |
<H1> Bayesian Optimization for Hyper-Parameter </H1> |
<H1> Gradient-Free Optimization for GLMNET Parameters </H1> |
<H1> Direct Optimization of Hyper-Parameter </H1> |
<H1> Sobol Sequence vs. Uniform Random in Hyper-Parameter Optimization </H1> |
<H1> Co-integration and Mean Reverting Portfolio </H1> |
<H1> Statistical Assessments of AUC </H1> |
<H1> Phillips-Ouliaris Test For Cointegration </H1> |
<H1> An Utility Function For Monotonic Binning </H1> |
<H1> Improving Binning by Bootstrap Bumping </H1> |
<H1> More Robust Monotonic Binning Based on Isotonic Regression </H1> |
<H1> Creating List with Iterator </H1> |
<H1> XFrames: Another Convenient Python Interface to Spark </H1> |
<H1> Growing List vs Growing Queue </H1> |
<H1> Convert Data Frame to Dictionary List in R </H1> |
<H1> Fetching Data From SAS Dataset to Lua Table </H1> |
<H1> Data Wrangling with Astropy Package </H1> |
<H1> Manipulating Dictionary List with SQLite Back-End </H1> |
<H1> Joining Dictionary Lists by Key </H1> |
<H1> Aggregation of Dictionary List by Key </H1> |
<H1> Convert SAS Dataset to Dictionary List </H1> |
<H1> Grouping Dictionary List by Key </H1> |
<H1> Subset Dictionary by Values </H1> |
<H1> Subset Dictionary by Keys </H1> |
<H1> Import CSV as Dictionary List </H1> |
<H1> Monotonic Binning with Equal-Sized Bads for Scorecard Development </H1> |
<H1> Two-Stage Estimation of Switching Regression </H1> |
<H1> By-Group Summary with SparkR – Follow-up for A Reader Comment </H1> |
<H1> Union Multiple Data.Frames with Different Column names </H1> |
<H1> Why Vectorize? </H1> |
<H1> How to Avoid For Loop in R </H1> |
<H1> Modeling Frequency Outcomes with Ordinal Models </H1> |
<H1> Playing Map() and Reduce() in R – Subsetting </H1> |
<H1> Playing Map() and Reduce() in R – By-Group Calculation </H1> |
<H1> Writing Wrapper in SAS </H1> |
<H1> More Flexible Ordinal Outcome Models </H1> |
<H1> Adjacent-Categories and Continuation-Ratio Logit Models for Ordinal Outcomes </H1> |
<H1> Ordered Probit Model and Price Movements of High-Frequency Trades </H1> |
<H1> Co-integration and Pairs Trading </H1> |
<H1> Subset by Index in Clojure </H1> |
<H1> SAS Implementation of ZAGA Models </H1> |
<H1> Mimicking SQLDF with MonetDBLite </H1> |
<H1> MLE with General Optimization Functions in R </H1> |
<H1> Read Random Rows from A Huge CSV File </H1> |
<H1> Updating Column Values in Clojure Map </H1> |
<H1> Adding New Columns to Clojure Map </H1> |
<H1> LogRatio Regression – A Simple Way to Model Compositional Data </H1> |
<H1> Transpose in Clojure </H1> |
<H1> Clojure Integration with R </H1> |
<H1> Aggregation by Multiple Keys in Clojure </H1> |
<H1> Inner and Outer Joins in Clojure </H1> |
<H1> By-Group Statistical Summary in Clojure </H1> |
<H1> Subset by Values in Clojure </H1> |
<H1> Do We Really Need Dataframe in Clojure? </H1> |
<H1> Parse CSV File with Headers in Clojure </H1> |
<H1> For Loop and Map in Clojure </H1> |
<H1> Clojure and SQLite </H1> |
<H1> MLE in R </H1> |
<H1> Modeling Dollar Amounts in Regression Setting </H1> |
<H1> R Interfaces to Python Keras Package </H1> |
<H1> Additional Thoughts on Estimating LGD with Proportional Odds Model </H1> |
<H1> Estimating Parameters of A Hyper-Poisson Distribution in SAS </H1> |
<H1> Modeling LGD with Proportional Odds Model </H1> |
<H1> Query CSV Data with Apache Drill </H1> |
<H1> Monotonic WoE Binning for LGD Models </H1> |
<H1> Granular Monotonic Binning in SAS </H1> |
<H1> Model Non-Negative Numeric Outcomes with Zeros </H1> |
<H1> Variable Selection with Elastic Net </H1> |
<H1> DART: Dropout Regularization in Boosting Ensembles </H1> |
<H1> Model Operational Losses with Copula Regression </H1> |
<H1> Sparkling Water and Moving Data Around </H1> |
<H1> Model Operational Loss Directly with Tweedie GLM </H1> |
<H1> GLM with H2O in R </H1> |
<H1> H2O Benchmark for CSV Import </H1> |
<H1> Using Tweedie Parameter to Identify Distributions </H1> |
<H1> Finer Monotonic Binning Based on Isotonic Regression </H1> |
<H1> Joining Tables in SparkR </H1> |
<H1> R Interface to Spark </H1> |
<H1> Data Aggregation with PySpark </H1> |
<H1> Kick Off Spark </H1> |
<H1> Double Poisson Regression in SAS </H1> |
<H1> SAS Macro Calculating Goodness-of-Fit Statistics for Quantile Regression </H1> |
<H1> Random Search for Optimal Parameters </H1> |
<H1> A Simple Convolutional Neural Network for The Binary Outcome </H1> |
<H1> Modeling Generalized Poisson Regression in SAS </H1> |
<H1> Monotonic Binning with Smbinning Package </H1> |
<H1> Autoencoder for Dimensionality Reduction </H1> |
<H1> An Example of Merge Layer in Keras </H1> |
<H1> Dropout Regularization in Deep Neural Networks </H1> |
<H1> Estimate Regression with (Type-I) Pareto Response </H1> |
<H1> Pregibon Test for Goodness of Link in SAS </H1> |
<H1> More about Flexible Frequency Models </H1> |
<H1> Modified Park Test in SAS </H1> |
<H1> Parameter Estimation of Pareto Type II Distribution with NLMIXED in SAS </H1> |
<H1> Fastest Way to Add New Variables to A Large Data.Frame </H1> |
<H1> Flavors of SQL on Pandas DataFrame </H1> |
<H1> Test Drive Proc Lua – Convert SAS Table to 2-Dimension Lua Table </H1> |
<H1> Copas Test for Overfitting in SAS </H1> |
<H1> SAS Macro Calculating Mutual Information </H1> |
<H1> Scorecard Development with Data from Multiple Sources </H1> |
<H1> Risk Models with Generalized PLS </H1> |
<H1> Duplicate Breusch-Godfrey Test Logic in SAS Autoreg Procedure </H1> |
<H1> More Flexible Approaches to Model Frequency </H1> |
<H1> Calculating ACF with Data Step Only </H1> |
<H1> Estimate Quasi-Binomial Model with GENMOD Procedure in SAS </H1> |
<H1> A More Flexible Ljung-Box Test in SAS </H1> |
<H1> SAS Macro Performing Breusch–Godfrey Test for Serial Correlation </H1> |
<H1> Python Prototype of Grid Search for SVM Parameters </H1> |
<H1> Improve SVM Tuning through Parallelism </H1> |
<H1> SAS Macro Calculating LOO Predictions for GLM </H1> |
<H1> Where Bagging Might Work Better Than Boosting </H1> |
<H1> The Power of Decision Stumps </H1> |
<H1> Parallelize Map() </H1> |
<H1> Parallelism with Joblib Package in Python </H1> |
<H1> Import CSV by Chunk Simultaneously with IPython Parallel </H1> |
<H1> Prediction Intervals for Poisson Regression </H1> |
<H1> Calculate Leave-One-Out Prediction for GLM </H1> |
<H1> Multivariate Adaptive Regression Splines with Python </H1> |
<H1> Download Federal Reserve Economic Data (FRED) with Python </H1> |
<H1> Fitting Generalized Regression Neural Network with Python </H1> |
<H1> Modeling Frequency in Operational Losses with Python </H1> |
<H1> Modeling Severity in Operational Losses with Python </H1> |
<H1> Posts navigation </H1> |
<H2> </H2> |
<H2> </H2> |
<H2> </H2> |
<H3> </H3> |
Social
Social Data
Cost and overhead previously rendered this semi-public form of communication unfeasible.
But advances in social networking technology from 2004-2010 has made broader concepts of sharing possible.