PREDICTION INTERVALS IN MACHINE LEARNING: RESIDUAL BOOTSTRAP AND QUANTILE REGRESSION FOR CASH FLOW ANALYSIS

Time series forecasting often faces challenges in producing reliable predictions due to inherent uncertainty in dynamic systems. While point predictions are commonly used, they may not adequately capture this uncertainty, especially in financial systems where forecasting accuracy directly impacts de...

Full description

Saved in:
Bibliographic Details
Main Authors: Wa Ode Rahmalia Safitri, Farit Mochamad Afendi, Budi Susetyo
Format: Article
Language:English
Published: Universitas Pattimura 2025-07-01
Series:Barekeng
Subjects:
Online Access:https://ojs3.unpatti.ac.id/index.php/barekeng/article/view/16433
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849237900464488448
author Wa Ode Rahmalia Safitri
Farit Mochamad Afendi
Budi Susetyo
author_facet Wa Ode Rahmalia Safitri
Farit Mochamad Afendi
Budi Susetyo
author_sort Wa Ode Rahmalia Safitri
collection DOAJ
description Time series forecasting often faces challenges in producing reliable predictions due to inherent uncertainty in dynamic systems. While point predictions are commonly used, they may not adequately capture this uncertainty, especially in financial systems where forecasting accuracy directly impacts decision-making. Prediction intervals offer a solution by providing a range of likely outcomes rather than single-point estimates. This study implements multivariate time series forecasting using gradient boosting algorithms (XGBoost, CatBoost, and LightGBM) to predict cash flow patterns in banking transactions, focusing on constructing reliable prediction intervals. Using transaction data from Bank Rakyat Indonesia (BRI), the research analyzes both office and e-channel transactions with different lag structures based on Granger Causality tests. Model performance was evaluated using RMSLE, MAE, and MAPE metrics, with RMSLE chosen as primary due to its ability to handle skewed distributions. LightGBM achieved best performance for office cash-in transactions (RMSLE: 0.2395), while CatBoost outperformed others for office cash-out (RMSLE: 0.2848), e-channel cash-in (RMSLE: 0.3946), and e-channel cash-out (RMSLE: 0.4221). For prediction intervals, two methods were compared: Residual Bootstrap with 500 samples and Quantile Regression. Residual Bootstrap generally produced coverage probabilities closer to the 80% level (i.e., 10–90% prediction interval), especially for office transactions, while maintaining narrower interval widths. In contrast, Quantile Regression tended to generate wider intervals and often overestimated uncertainty, resulting in overly high coverage in some cases. However, both methods showed clear limitations when applied to e-channel transactions, particularly for cash-in e-channel, where coverage probabilities fell below 50% due to high volatility and irregular transaction patterns. Unlike previous work focused only on point forecasts, this study offers insights into forecast uncertainty by evaluating how well each method quantifies, providing practical guidance for financial institutions aiming to improve risk management through interval-based forecasting.
format Article
id doaj-art-2a23d2ba21ff47288f8c3e459acebe21
institution Kabale University
issn 1978-7227
2615-3017
language English
publishDate 2025-07-01
publisher Universitas Pattimura
record_format Article
series Barekeng
spelling doaj-art-2a23d2ba21ff47288f8c3e459acebe212025-08-20T04:01:48ZengUniversitas PattimuraBarekeng1978-72272615-30172025-07-011931625163610.30598/barekengvol19iss3pp1625-163616433PREDICTION INTERVALS IN MACHINE LEARNING: RESIDUAL BOOTSTRAP AND QUANTILE REGRESSION FOR CASH FLOW ANALYSISWa Ode Rahmalia Safitri0Farit Mochamad Afendi1Budi Susetyo2Statistics and Data Science Department, Faculty of Mathematics and Natural Sciences, IPB University, IndonesiaStatistics and Data Science Department, Faculty of Mathematics and Natural Sciences, IPB University, IndonesiaStatistics and Data Science Department, Faculty of Mathematics and Natural Sciences, IPB University, IndonesiaTime series forecasting often faces challenges in producing reliable predictions due to inherent uncertainty in dynamic systems. While point predictions are commonly used, they may not adequately capture this uncertainty, especially in financial systems where forecasting accuracy directly impacts decision-making. Prediction intervals offer a solution by providing a range of likely outcomes rather than single-point estimates. This study implements multivariate time series forecasting using gradient boosting algorithms (XGBoost, CatBoost, and LightGBM) to predict cash flow patterns in banking transactions, focusing on constructing reliable prediction intervals. Using transaction data from Bank Rakyat Indonesia (BRI), the research analyzes both office and e-channel transactions with different lag structures based on Granger Causality tests. Model performance was evaluated using RMSLE, MAE, and MAPE metrics, with RMSLE chosen as primary due to its ability to handle skewed distributions. LightGBM achieved best performance for office cash-in transactions (RMSLE: 0.2395), while CatBoost outperformed others for office cash-out (RMSLE: 0.2848), e-channel cash-in (RMSLE: 0.3946), and e-channel cash-out (RMSLE: 0.4221). For prediction intervals, two methods were compared: Residual Bootstrap with 500 samples and Quantile Regression. Residual Bootstrap generally produced coverage probabilities closer to the 80% level (i.e., 10–90% prediction interval), especially for office transactions, while maintaining narrower interval widths. In contrast, Quantile Regression tended to generate wider intervals and often overestimated uncertainty, resulting in overly high coverage in some cases. However, both methods showed clear limitations when applied to e-channel transactions, particularly for cash-in e-channel, where coverage probabilities fell below 50% due to high volatility and irregular transaction patterns. Unlike previous work focused only on point forecasts, this study offers insights into forecast uncertainty by evaluating how well each method quantifies, providing practical guidance for financial institutions aiming to improve risk management through interval-based forecasting.https://ojs3.unpatti.ac.id/index.php/barekeng/article/view/16433bootstrap methodcatboostlightgbmprediction intervalsquantile regressionxgboost
spellingShingle Wa Ode Rahmalia Safitri
Farit Mochamad Afendi
Budi Susetyo
PREDICTION INTERVALS IN MACHINE LEARNING: RESIDUAL BOOTSTRAP AND QUANTILE REGRESSION FOR CASH FLOW ANALYSIS
Barekeng
bootstrap method
catboost
lightgbm
prediction intervals
quantile regression
xgboost
title PREDICTION INTERVALS IN MACHINE LEARNING: RESIDUAL BOOTSTRAP AND QUANTILE REGRESSION FOR CASH FLOW ANALYSIS
title_full PREDICTION INTERVALS IN MACHINE LEARNING: RESIDUAL BOOTSTRAP AND QUANTILE REGRESSION FOR CASH FLOW ANALYSIS
title_fullStr PREDICTION INTERVALS IN MACHINE LEARNING: RESIDUAL BOOTSTRAP AND QUANTILE REGRESSION FOR CASH FLOW ANALYSIS
title_full_unstemmed PREDICTION INTERVALS IN MACHINE LEARNING: RESIDUAL BOOTSTRAP AND QUANTILE REGRESSION FOR CASH FLOW ANALYSIS
title_short PREDICTION INTERVALS IN MACHINE LEARNING: RESIDUAL BOOTSTRAP AND QUANTILE REGRESSION FOR CASH FLOW ANALYSIS
title_sort prediction intervals in machine learning residual bootstrap and quantile regression for cash flow analysis
topic bootstrap method
catboost
lightgbm
prediction intervals
quantile regression
xgboost
url https://ojs3.unpatti.ac.id/index.php/barekeng/article/view/16433
work_keys_str_mv AT waoderahmaliasafitri predictionintervalsinmachinelearningresidualbootstrapandquantileregressionforcashflowanalysis
AT faritmochamadafendi predictionintervalsinmachinelearningresidualbootstrapandquantileregressionforcashflowanalysis
AT budisusetyo predictionintervalsinmachinelearningresidualbootstrapandquantileregressionforcashflowanalysis