{ "_id": "6a1edd88b401979e73410124", "Package": "EQRN", "Type": "Package", "Title": "Extreme Quantile Regression Neural Networks for Risk Forecasting", "Version": "0.1.2", "Authors@R": "c(person(c(\"Olivier\", \"C.\"), \"Pasche\",\nemail = \"olivier_pasche@alumni.epfl.ch\", role = c(\"aut\", \"cre\", \"cph\"),\ncomment = c(ORCID = \"0000-0002-1202-9199\"))\n)", "Description": "This framework enables forecasting and extrapolating\nmeasures of conditional risk (e.g. of extreme or unprecedented\nevents), including quantiles and exceedance probabilities,\nusing extreme value statistics and flexible neural network\narchitectures. It allows for capturing complex multivariate\ndependencies, including dependencies between observations, such\nas sequential dependence (time-series). The methodology was\nintroduced in Pasche and Engelke (2024)\n (also available in preprint: Pasche\nand Engelke (2022) ).", "License": "GPL (>= 3)", "Encoding": "UTF-8", "Roxygen": "list(markdown = TRUE)", "RoxygenNote": "7.3.2", "URL": "https://github.com/opasche/EQRN, https://opasche.github.io/EQRN/", "BugReports": "https://github.com/opasche/EQRN/issues", "Repository": "https://opasche.r-universe.dev", "Date/Publication": "2025-11-21 13:41:07 UTC", "RemoteUrl": "https://github.com/opasche/eqrn", "RemoteRef": "HEAD", "RemoteSha": "c3b007e747644b896b22c8eb1cdffb1976d5f63d", "NeedsCompilation": "no", "Packaged": { "Date": "2026-05-20 08:12:00 UTC", "User": "root" }, "Author": "Olivier C. Pasche [aut, cre, cph] (ORCID:\n)", "Maintainer": "Olivier C. Pasche ", "MD5sum": "5206fc3e57ee4277aff87b29546a552a", "_user": "opasche", "_type": "src", "_file": "EQRN_0.1.2.tar.gz", "_fileid": "5f840c601ab3a03065311ae1cd38c84e4f15b63b5a4dbcde8a88a28e1d695870", "_filesize": 302395, "_sha256": "5f840c601ab3a03065311ae1cd38c84e4f15b63b5a4dbcde8a88a28e1d695870", "_created": "2026-05-20T08:12:00.000Z", "_published": "2026-06-02T13:41:28.582Z", "_distro": "noble", "_jobs": [ { "job": 79084458938, "time": 150, "config": "linux-devel-x86_64", "r": "4.7.0", "check": "OK", "artifact": "7104690560" }, { "job": 79084459015, "time": 149, "config": "linux-release-x86_64", "r": "4.6.0", "check": "OK", "artifact": "7104690133" }, { "job": 79084458877, "time": 103, "config": "macos-oldrel-arm64", "r": "4.5.3", "check": "OK", "artifact": "7104673783" }, { "job": 79084458929, "time": 100, "config": "macos-release-arm64", "r": "4.6.0", "check": "OK", "artifact": "7104673051" }, { "job": 79084457747, "time": 172, "config": "source", "r": "4.6.0", "check": "OK", "artifact": "7104637807" }, { "job": 79084457892, "time": 112, "config": "wasm-release", "r": "4.6.0", "check": "OK", "artifact": "7359366175" }, { "job": 79084458947, "time": 163, "config": "windows-devel", "r": "4.7.0", "check": "OK", "artifact": "7104694520" }, { "job": 79084459099, "time": 116, "config": "windows-oldrel", "r": "4.5.3", "check": "OK", "artifact": "7104680229" }, { "job": 79084459171, "time": 103, "config": "windows-release", "r": "4.6.0", "check": "OK", "artifact": "7104674457" } ], "_buildurl": "https://github.com/r-universe/opasche/actions/runs/26149934342", "_status": "success", "_host": "GitHub-Actions", "_upstream": "https://github.com/opasche/eqrn", "_commit": { "id": "c3b007e747644b896b22c8eb1cdffb1976d5f63d", "author": "OPasche ", "committer": "OPasche ", "message": "New CRAN patch version (fixes #1)\n", "time": 1763732467 }, "_maintainer": { "name": "Olivier C. Pasche", "email": "olivier_pasche@alumni.epfl.ch", "login": "opasche", "orcid": "0000-0002-1202-9199", "description": "Lecturer in Statistics (Ph.D.)", "uuid": 32179312 }, "_registered": true, "_dependencies": [ { "package": "coro", "version": ">= 1.0.2", "role": "Imports" }, { "package": "doFuture", "role": "Imports" }, { "package": "evd", "role": "Imports" }, { "package": "foreach", "role": "Imports" }, { "package": "future", "role": "Imports" }, { "package": "ismev", "role": "Imports" }, { "package": "magrittr", "role": "Imports" }, { "package": "stats", "role": "Imports" }, { "package": "torch", "role": "Imports" }, { "package": "utils", "role": "Imports" } ], "_owner": "opasche", "_selfowned": true, "_usedby": 0, "_updates": [ { "week": "2025-44", "n": 3 }, { "week": "2025-46", "n": 2 }, { "week": "2025-47", "n": 1 } ], "_tags": [ { "name": "v0.1.2", "date": "2025-11-21" } ], "_stars": 18, "_contributors": [ { "user": "opasche", "count": 22, "uuid": 32179312 } ], "_userbio": { "uuid": 32179312, "type": "user", "name": "OPasche", "description": "Lecturer in Statistics (Ph.D.)" }, "_downloads": { "count": 623, "source": "https://cranlogs.r-pkg.org/downloads/total/last-month/EQRN" }, "_devurl": "https://github.com/opasche/eqrn", "_pkgdown": "https://opasche.github.io/EQRN/", "_searchresults": 13, "_rbuild": "4.6.0", "_assets": [ "extra/citation.cff", "extra/citation.html", "extra/citation.json", "extra/citation.txt", "extra/contents.json", "extra/EQRN.html", "extra/NEWS.html", "extra/NEWS.txt", "extra/readme.html", "extra/readme.md", "manual.pdf" ], "_homeurl": "https://github.com/opasche/eqrn", "_realowner": "opasche", "_cranurl": true, "_releases": [ { "version": "0.1.1", "date": "2025-03-17" }, { "version": "0.1.2", "date": "2025-11-21" } ], "_exports": [ "backend_is_installed", "check_directory", "compute_EQRN_GPDLoss", "compute_EQRN_seq_GPDLoss", "default_device", "end_doFuture_strategy", "EQRN_excess_probability", "EQRN_excess_probability_seq", "EQRN_fit", "EQRN_fit_restart", "EQRN_fit_seq", "EQRN_load", "EQRN_predict", "EQRN_predict_params", "EQRN_predict_params_seq", "EQRN_predict_seq", "EQRN_save", "excess_probability", "FC_GPD_net", "FC_GPD_SNN", "fit_GPD_unconditional", "get_doFuture_operator", "get_excesses", "GPD_excess_probability", "GPD_quantiles", "install_backend", "lagged_features", "last_elem", "loss_GPD", "loss_GPD_tensor", "make_folds", "mean_absolute_error", "mean_squared_error", "mts_dataset", "multilevel_exceedance_proba_error", "multilevel_MAE", "multilevel_MSE", "multilevel_pred_bias", "multilevel_prop_below", "multilevel_q_loss", "multilevel_q_pred_error", "multilevel_R_squared", "multilevel_resid_var", "perform_scaling", "predict_GPD_semiconditional", "predict_unconditional_quantiles", "prediction_bias", "prediction_residual_variance", "process_features", "proportion_below", "QRN_fit_multiple", "QRN_seq_fit", "QRN_seq_predict", "QRN_seq_predict_foldwise", "QRN_seq_predict_foldwise_sep", "QRNN_RNN_net", "quantile_exceedance_proba_error", "quantile_loss", "quantile_loss_tensor", "quantile_prediction_error", "R_squared", "Recurrent_GPD_net", "roundm", "safe_save_rds", "semiconditional_train_valid_GPD_loss", "Separated_GPD_SNN", "set_doFuture_strategy", "square_loss", "unconditional_train_valid_GPD_loss", "vec2mat", "vector_insert" ], "_help": [ { "page": "backend_is_installed", "title": "Check if Torch Backend Libraries are Installed", "topics": [ "backend_is_installed" ] }, { "page": "check_directory", "title": "Check directory existence", "topics": [ "check_directory" ] }, { "page": "compute_EQRN_GPDLoss", "title": "Generalized Pareto likelihood loss of a EQRN_iid predictor", "topics": [ "compute_EQRN_GPDLoss" ] }, { "page": "compute_EQRN_seq_GPDLoss", "title": "Generalized Pareto likelihood loss of a EQRN_seq predictor", "topics": [ "compute_EQRN_seq_GPDLoss" ] }, { "page": "default_device", "title": "Default torch device", "topics": [ "default_device" ] }, { "page": "end_doFuture_strategy", "title": "End the currently set doFuture strategy", "topics": [ "end_doFuture_strategy" ] }, { "page": "EQRN_excess_probability", "title": "Tail excess probability prediction using an EQRN_iid object", "topics": [ "EQRN_excess_probability" ] }, { "page": "EQRN_excess_probability_seq", "title": "Tail excess probability prediction using an EQRN_seq object", "topics": [ "EQRN_excess_probability_seq" ] }, { "page": "EQRN_fit", "title": "EQRN fit function for independent data", "topics": [ "EQRN_fit" ] }, { "page": "EQRN_fit_restart", "title": "Wrapper for fitting EQRN with restart for stability", "topics": [ "EQRN_fit_restart" ] }, { "page": "EQRN_fit_seq", "title": "EQRN fit function for sequential and time series data", "topics": [ "EQRN_fit_seq" ] }, { "page": "EQRN_load", "title": "Load an EQRN object from disc", "topics": [ "EQRN_load" ] }, { "page": "EQRN_predict", "title": "Predict function for an EQRN_iid fitted object", "topics": [ "EQRN_predict" ] }, { "page": "EQRN_predict_params", "title": "GPD parameters prediction function for an EQRN_iid fitted object", "topics": [ "EQRN_predict_params" ] }, { "page": "EQRN_predict_params_seq", "title": "GPD parameters prediction function for an EQRN_seq fitted object", "topics": [ "EQRN_predict_params_seq" ] }, { "page": "EQRN_predict_seq", "title": "Predict function for an EQRN_seq fitted object", "topics": [ "EQRN_predict_seq" ] }, { "page": "EQRN_save", "title": "Save an EQRN object on disc", "topics": [ "EQRN_save" ] }, { "page": "excess_probability", "title": "Excess Probability Predictions", "topics": [ "excess_probability" ] }, { "page": "excess_probability.EQRN_iid", "title": "Tail excess probability prediction method using an EQRN_iid object", "topics": [ "excess_probability.EQRN_iid" ] }, { "page": "excess_probability.EQRN_seq", "title": "Tail excess probability prediction method using an EQRN_iid object", "topics": [ "excess_probability.EQRN_seq" ] }, { "page": "FC_GPD_net", "title": "MLP module for GPD parameter prediction", "topics": [ "FC_GPD_net" ] }, { "page": "FC_GPD_SNN", "title": "Self-normalized fully-connected network module for GPD parameter prediction", "topics": [ "FC_GPD_SNN" ] }, { "page": "fit_GPD_unconditional", "title": "Maximum likelihood estimates for the GPD distribution using peaks over threshold", "topics": [ "fit_GPD_unconditional" ] }, { "page": "get_doFuture_operator", "title": "Get doFuture operator", "topics": [ "get_doFuture_operator" ] }, { "page": "get_excesses", "title": "Computes rescaled excesses over the conditional quantiles", "topics": [ "get_excesses" ] }, { "page": "GPD_excess_probability", "title": "Tail excess probability prediction based on conditional GPD parameters", "topics": [ "GPD_excess_probability" ] }, { "page": "GPD_quantiles", "title": "Compute extreme quantile from GPD parameters", "topics": [ "GPD_quantiles" ] }, { "page": "install_backend", "title": "Install Torch Backend Libraries", "topics": [ "install_backend" ] }, { "page": "lagged_features", "title": "Covariate lagged replication for temporal dependence", "topics": [ "lagged_features" ] }, { "page": "last_elem", "title": "Last element of a vector", "topics": [ "last_elem" ] }, { "page": "loss_GPD", "title": "Generalized Pareto likelihood loss", "topics": [ "loss_GPD" ] }, { "page": "loss_GPD_tensor", "title": "GPD tensor loss function for training a EQRN network", "topics": [ "loss_GPD_tensor" ] }, { "page": "make_folds", "title": "Create cross-validation folds", "topics": [ "make_folds" ] }, { "page": "mean_absolute_error", "title": "Mean absolute error", "topics": [ "mean_absolute_error" ] }, { "page": "mean_squared_error", "title": "Mean squared error", "topics": [ "mean_squared_error" ] }, { "page": "mts_dataset", "title": "Dataset creator for sequential data", "topics": [ "mts_dataset" ] }, { "page": "multilevel_exceedance_proba_error", "title": "Multilevel 'quantile_exceedance_proba_error'", "topics": [ "multilevel_exceedance_proba_error" ] }, { "page": "multilevel_MAE", "title": "Multilevel quantile MAEs", "topics": [ "multilevel_MAE" ] }, { "page": "multilevel_MSE", "title": "Multilevel quantile MSEs", "topics": [ "multilevel_MSE" ] }, { "page": "multilevel_pred_bias", "title": "Multilevel prediction bias", "topics": [ "multilevel_pred_bias" ] }, { "page": "multilevel_prop_below", "title": "Multilevel 'proportion_below'", "topics": [ "multilevel_prop_below" ] }, { "page": "multilevel_q_loss", "title": "Multilevel quantile losses", "topics": [ "multilevel_q_loss" ] }, { "page": "multilevel_q_pred_error", "title": "Multilevel 'quantile_prediction_error'", "topics": [ "multilevel_q_pred_error" ] }, { "page": "multilevel_R_squared", "title": "Multilevel R squared", "topics": [ "multilevel_R_squared" ] }, { "page": "multilevel_resid_var", "title": "Multilevel residual variance", "topics": [ "multilevel_resid_var" ] }, { "page": "perform_scaling", "title": "Performs feature scaling without overfitting", "topics": [ "perform_scaling" ] }, { "page": "predict_GPD_semiconditional", "title": "Predict semi-conditional extreme quantiles using peaks over threshold", "topics": [ "predict_GPD_semiconditional" ] }, { "page": "predict_unconditional_quantiles", "title": "Predict unconditional extreme quantiles using peaks over threshold", "topics": [ "predict_unconditional_quantiles" ] }, { "page": "predict.EQRN_iid", "title": "Predict method for an EQRN_iid fitted object", "topics": [ "predict.EQRN_iid" ] }, { "page": "predict.EQRN_seq", "title": "Predict method for an EQRN_seq fitted object", "topics": [ "predict.EQRN_seq" ] }, { "page": "predict.QRN_seq", "title": "Predict method for a QRN_seq fitted object", "topics": [ "predict.QRN_seq" ] }, { "page": "prediction_bias", "title": "Prediction bias", "topics": [ "prediction_bias" ] }, { "page": "prediction_residual_variance", "title": "Prediction residual variance", "topics": [ "prediction_residual_variance" ] }, { "page": "process_features", "title": "Feature processor for EQRN", "topics": [ "process_features" ] }, { "page": "proportion_below", "title": "Proportion of observations below conditional quantile vector", "topics": [ "proportion_below" ] }, { "page": "QRN_fit_multiple", "title": "Wrapper for fitting a recurrent QRN with restart for stability", "topics": [ "QRN_fit_multiple" ] }, { "page": "QRN_seq_fit", "title": "Recurrent QRN fitting function", "topics": [ "QRN_seq_fit" ] }, { "page": "QRN_seq_predict", "title": "Predict function for a QRN_seq fitted object", "topics": [ "QRN_seq_predict" ] }, { "page": "QRN_seq_predict_foldwise", "title": "Foldwise fit-predict function using a recurrent QRN", "topics": [ "QRN_seq_predict_foldwise" ] }, { "page": "QRN_seq_predict_foldwise_sep", "title": "Sigle-fold foldwise fit-predict function using a recurrent QRN", "topics": [ "QRN_seq_predict_foldwise_sep" ] }, { "page": "QRNN_RNN_net", "title": "Recurrent quantile regression neural network module", "topics": [ "QRNN_RNN_net" ] }, { "page": "quantile_exceedance_proba_error", "title": "Quantile exceedance probability prediction calibration error", "topics": [ "quantile_exceedance_proba_error" ] }, { "page": "quantile_loss", "title": "Quantile loss", "topics": [ "quantile_loss" ] }, { "page": "quantile_loss_tensor", "title": "Tensor quantile loss function for training a QRN network", "topics": [ "quantile_loss_tensor" ] }, { "page": "quantile_prediction_error", "title": "Quantile prediction calibration error", "topics": [ "quantile_prediction_error" ] }, { "page": "R_squared", "title": "R squared", "topics": [ "R_squared" ] }, { "page": "Recurrent_GPD_net", "title": "Recurrent network module for GPD parameter prediction", "topics": [ "Recurrent_GPD_net" ] }, { "page": "roundm", "title": "Mathematical number rounding", "topics": [ "roundm" ] }, { "page": "safe_save_rds", "title": "Safe RDS save", "topics": [ "safe_save_rds" ] }, { "page": "semiconditional_train_valid_GPD_loss", "title": "Semi-conditional GPD MLEs and their train-validation likelihoods", "topics": [ "semiconditional_train_valid_GPD_loss" ] }, { "page": "Separated_GPD_SNN", "title": "Self-normalized separated network module for GPD parameter prediction", "topics": [ "Separated_GPD_SNN" ] }, { "page": "set_doFuture_strategy", "title": "Set a doFuture execution strategy", "topics": [ "set_doFuture_strategy" ] }, { "page": "square_loss", "title": "Square loss", "topics": [ "square_loss" ] }, { "page": "unconditional_train_valid_GPD_loss", "title": "Unconditional GPD MLEs and their train-validation likelihoods", "topics": [ "unconditional_train_valid_GPD_loss" ] }, { "page": "vec2mat", "title": "Convert a vector to a matrix", "topics": [ "vec2mat" ] }, { "page": "vector_insert", "title": "Insert value in vector", "topics": [ "vector_insert" ] } ], "_readme": "https://github.com/opasche/eqrn/raw/HEAD/README.md", "_rundeps": [ "bit", "bit64", "callr", "cli", "codetools", "coro", "desc", "digest", "doFuture", "evd", "farver", "foreach", "future", "future.apply", "globals", "glue", "ismev", "iterators", "jsonlite", "labeling", "lattice", "lifecycle", "listenv", "magrittr", "Matrix", "mgcv", "nlme", "parallelly", "processx", "ps", "R6", "RColorBrewer", "Rcpp", "rlang", "safetensors", "scales", "torch", "viridisLite", "withr" ], "_score": 4.545307116465825, "_indexed": true, "_nocasepkg": "eqrn", "_universes": [ "opasche" ], "_binaries": [ { "r": "4.7.0", "os": "linux", "version": "0.1.2", "date": "2026-05-20T08:14:10.000Z", "distro": "noble", "commit": "c3b007e747644b896b22c8eb1cdffb1976d5f63d", "fileid": "617e6f9ad99b796231e622c4e96b5961ee0ad57bb7c6fbeda1ac6aede6471a74", "status": "success", "check": "OK", "buildurl": "https://github.com/r-universe/opasche/actions/runs/26149934342" }, { "r": "4.6.0", "os": "linux", "version": "0.1.2", "date": "2026-05-20T08:14:07.000Z", "distro": "noble", "commit": "c3b007e747644b896b22c8eb1cdffb1976d5f63d", "fileid": "cde47fc9ff83197331ec8d4e79d479097dce03dd972108f46304169e365d1e7b", "status": "success", "check": "OK", "buildurl": "https://github.com/r-universe/opasche/actions/runs/26149934342" }, { "r": "4.5.3", "os": "mac", "version": "0.1.2", "date": "2026-05-20T08:13:37.000Z", "commit": "c3b007e747644b896b22c8eb1cdffb1976d5f63d", "fileid": "c1718cc9c730861fcb73659c3775ff642b2199e23a0b9512b299c58bfbab93d8", "status": "success", "check": "OK", "buildurl": "https://github.com/r-universe/opasche/actions/runs/26149934342" }, { "r": "4.6.0", "os": "mac", "version": "0.1.2", "date": "2026-05-20T08:13:35.000Z", "commit": "c3b007e747644b896b22c8eb1cdffb1976d5f63d", "fileid": "5976a20daa3c0a621c961d7187eff738ab3d50b695288816f7d16e69d73a3bd6", "status": "success", "check": "OK", "buildurl": "https://github.com/r-universe/opasche/actions/runs/26149934342" }, { "r": "4.7.0", "os": "win", "version": "0.1.2", "date": "2026-05-20T08:14:15.000Z", "commit": "c3b007e747644b896b22c8eb1cdffb1976d5f63d", "fileid": "707e54db6a9f093981700c4e3f88e0baf05288aea9d1e273916ae9b61761f04c", "status": "success", "check": "OK", "buildurl": "https://github.com/r-universe/opasche/actions/runs/26149934342" }, { "r": "4.5.3", "os": "win", "version": "0.1.2", "date": "2026-05-20T08:13:27.000Z", "commit": "c3b007e747644b896b22c8eb1cdffb1976d5f63d", "fileid": "ada31a761d71fe2f29c049917fb94caef5fb1abbf18cb851c1bf73bdf6b62e67", "status": "success", "check": "OK", "buildurl": "https://github.com/r-universe/opasche/actions/runs/26149934342" }, { "r": "4.6.0", "os": "win", "version": "0.1.2", "date": "2026-05-20T08:13:22.000Z", "commit": "c3b007e747644b896b22c8eb1cdffb1976d5f63d", "fileid": "b93806211d36b623336ecab282d0d119c032b25f91744913b6061ed91976e681", "status": "success", "check": "OK", "buildurl": "https://github.com/r-universe/opasche/actions/runs/26149934342" }, { "r": "4.6.0", "os": "wasm", "version": "0.1.2", "date": "2026-06-02T13:41:06.000Z", "commit": "c3b007e747644b896b22c8eb1cdffb1976d5f63d", "fileid": "63a9dbef2f7ac0e3b40e63143eb78384f8b581087de6af749818c557fb528f68", "status": "success", "buildurl": "https://github.com/r-universe/opasche/actions/runs/26149934342" } ] }