||
[资料] 时间序列分析与预测的常用误差统计指标
平均误差:ME (mean error)
平均百分比误差:MPE (Mean Percentage Error)
平均绝对比例误差:MASE (Mean Absolute Scaled Error)
平均绝对误差:MAE (mean absolute error)
均方误差:MSE (mean squared error)
均方根误差:RMSE (root mean squared error)
平均绝对相对误差:MARE (mean absolute relative error)
均方相对误差:MSRE (mean squared relative error)
均方根相对误差:RMSRE (root mean squared relative error)
平均绝对百分比误差:MAPE (mean absolute percentage error)
均方误差百分比:MSPE (mean squared percentage error)
均方根百分比误差:RMSPE (root mean squared percentage error)
最大误差:Emax (误差绝对值的最大值, maximum of absolute value of error)。尽管这个量不是统计量。
陆续抄袭来的误差计算公式:
感谢有关人员!
傻以为:对于时间序列预测,平均误差ME、均方相对误差RMSE、最大误差Emax,是三个必须的误差值。
参考资料:
[1] 久未见, 2020-03-07, 评价指标RMSE、MSE、MAE、MAPE、SMAPE 、R-Squared——python+sklearn实现
https://www.e-learn.cn/topic/3481165
[2] 手撕机, 2019-02-21, 预测评价指标RMSE、MSE、MAE、MAPE、SMAPE
https://blog.csdn.net/guolindonggld/article/details/87856780
https://www.cnblogs.com/think90/articles/11786016.html
[3] Data Science, MAD vs RMSE vs MAE vs MSLE vs R2: When to use which?
[4] MathWorks, Error related performance metrics
https://www.mathworks.com/matlabcentral/fileexchange/15130-error-related-performance-metrics
[5] Rob Hyndman, Accuracy measures for a forecast model
https://pkg.robjhyndman.com/forecast/reference/accuracy.html
[6] BL-Graphics, Forecast KPI: Bias, MAE, MAPE & RMSE
https://supchains.com/article/forecast-kpi-bias-mae-mape-rmse/
[7] Computer Age Statistical Inference: Algorithms, Evidence and Data Science
https://web.stanford.edu/~hastie/CASI/
相关链接:
[1] 2021-02-05,[笔记] 时间序列预测里的一些“专业术语”
http://blog.sciencenet.cn/blog-107667-1270762.html
[2] 2019-06-08,[学习资料搜集] 时间序列(time series)学习书籍
http://blog.sciencenet.cn/blog-107667-1183775.html
[3] 2018-08-18,“大数据”时期,更渴望“小样本数理统计学”
http://blog.sciencenet.cn/blog-107667-1129894.html
[4] 2016-09-01,Crosswavelet and Wavelet Coherence 小波分析的程序网址
http://blog.sciencenet.cn/blog-107667-1000091.html
[5] 2016-09-01,支持向量机 Support Vector Machine 程序网址
http://blog.sciencenet.cn/blog-107667-1000087.html
感谢您的指教!
感谢您指正以上任何错误!
感谢您提供更多的相关资料!
Archiver|手机版|科学网 ( 京ICP备07017567号-12 )
GMT+8, 2024-11-19 22:41
Powered by ScienceNet.cn
Copyright © 2007- 中国科学报社