Paul Bernal
2018-Jul-02 20:22 UTC
[R] Extracting the MAPE value from a fitted Time Series Model
Dear friends, I want to extract the MAPE value from a fitted time series model. This is what I have:> str(TransitSpline)List of 12 $ method : chr "Cubic Smoothing Spline" $ level : num [1:2] 80 95 $ x : Time-Series [1:385] from 1 to 385: 77 75 85 74 73 96 82 90 91 81 ... $ series : chr "data$Transits" $ mean : Time-Series [1:10, 1] from 386 to 395: 186 178 170 163 155 ... ..- attr(*, "dimnames")=List of 2 .. ..$ : NULL .. ..$ : chr "Series 1" $ upper : Time-Series [1:10, 1:2] from 386 to 395: 202 199 197 197 197 ... ..- attr(*, "dimnames")=List of 2 .. ..$ : NULL .. ..$ : chr [1:2] "Series 1" "Series 2" $ lower : Time-Series [1:10, 1:2] from 386 to 395: 171 158 144 129 113 ... ..- attr(*, "dimnames")=List of 2 .. ..$ : NULL .. ..$ : chr [1:2] "Series 1" "Series 2" $ model :List of 2 ..$ beta: num 6.15 ..$ call: language splinef(y = data$Transits) $ fitted : Time-Series [1:385] from 1 to 385: 76.1 77.3 78.5 80.1 82.2 ... $ residuals : Time-Series [1:385] from 1 to 385: NA -1.3 9.49 -8.64 -4.34 ... $ standardizedresiduals: Time-Series [1:385] from 1 to 385: NA -0.875 6.517 -5.586 -2.736 ... $ onestepf : Time-Series [1:385] from 1 to 385: NA 76.3 75.5 82.6 77.3 ... - attr(*, "class")= chr [1:2] "splineforecast" "forecast"> str(summary(TransitSpline))#Here I want to get the value for the MAPE measure Forecast method: Cubic Smoothing Spline Model Information: $`beta` [1] 6.149167 $call splinef(y = data$Transits) Error measures: ME RMSE MAE MPE MAPE MASE ACF1 Training set -0.07776434 12.10204 9.003675 -0.2408687 5.377131 0.930913 -0.2766975 Forecasts: Point Forecast Lo 80 Hi 80 Lo 95 Hi 95 386 186.0153 170.52426 201.5064 162.323777 209.7069 387 178.2220 157.87687 198.5671 147.106804 209.3372 388 170.4287 143.80863 197.0487 129.716832 211.1405 389 162.6353 128.61257 196.6581 110.602006 214.6687 390 154.8420 112.52646 197.1576 90.125956 219.5581 391 147.0487 95.66491 198.4324 68.463984 225.6334 392 139.2553 78.10706 200.4036 45.737114 232.7736 393 131.4620 59.92462 202.9994 22.055013 240.8690 394 123.6687 41.14798 206.1894 -2.535833 249.8732 395 115.8753 21.82457 209.9261 -27.962900 259.7136 'data.frame': 10 obs. of 5 variables: $ Point Forecast: num 186 178 170 163 155 ... $ Lo 80 : num 171 158 144 129 113 ... $ Hi 80 : num 202 199 197 197 197 ... $ Lo 95 : num 162.3 147.1 129.7 110.6 90.1 ... $ Hi 95 : num 210 209 211 215 220 ... any idea on how to accomplish this? Best regards, Paul [[alternative HTML version deleted]]
Jeff Newmiller
2018-Jul-02 20:40 UTC
[R] Extracting the MAPE value from a fitted Time Series Model
Google offers [1], which probably seems like a vague response but your question omitted a reproducible example and is contaminated by posting in HTML (read the Posting Guide). [1] https://www.rdocumentation.org/packages/MLmetrics/versions/1.1.1/topics/MAPE On July 2, 2018 1:22:39 PM PDT, Paul Bernal <paulbernal07 at gmail.com> wrote:>Dear friends, > >I want to extract the MAPE value from a fitted time series model. This >is >what I have: > >> str(TransitSpline) >List of 12 > $ method : chr "Cubic Smoothing Spline" > $ level : num [1:2] 80 95 >$ x : Time-Series [1:385] from 1 to 385: 77 75 85 74 >73 >96 82 90 91 81 ... > $ series : chr "data$Transits" >$ mean : Time-Series [1:10, 1] from 386 to 395: 186 178 >170 163 155 ... > ..- attr(*, "dimnames")=List of 2 > .. ..$ : NULL > .. ..$ : chr "Series 1" >$ upper : Time-Series [1:10, 1:2] from 386 to 395: 202 >199 >197 197 197 ... > ..- attr(*, "dimnames")=List of 2 > .. ..$ : NULL > .. ..$ : chr [1:2] "Series 1" "Series 2" >$ lower : Time-Series [1:10, 1:2] from 386 to 395: 171 >158 >144 129 113 ... > ..- attr(*, "dimnames")=List of 2 > .. ..$ : NULL > .. ..$ : chr [1:2] "Series 1" "Series 2" > $ model :List of 2 > ..$ beta: num 6.15 > ..$ call: language splinef(y = data$Transits) >$ fitted : Time-Series [1:385] from 1 to 385: 76.1 77.3 >78.5 >80.1 82.2 ... >$ residuals : Time-Series [1:385] from 1 to 385: NA -1.3 >9.49 >-8.64 -4.34 ... > $ standardizedresiduals: Time-Series [1:385] from 1 to 385: NA -0.875 >6.517 -5.586 -2.736 ... >$ onestepf : Time-Series [1:385] from 1 to 385: NA 76.3 >75.5 >82.6 77.3 ... > - attr(*, "class")= chr [1:2] "splineforecast" "forecast" > > >> str(summary(TransitSpline)) >#Here I want to get the value for the MAPE measure >Forecast method: Cubic Smoothing Spline > >Model Information: >$`beta` >[1] 6.149167 > >$call >splinef(y = data$Transits) > > >Error measures: > ME RMSE MAE MPE MAPE MASE > ACF1 >Training set -0.07776434 12.10204 9.003675 -0.2408687 5.377131 0.930913 >-0.2766975 > >Forecasts: > Point Forecast Lo 80 Hi 80 Lo 95 Hi 95 >386 186.0153 170.52426 201.5064 162.323777 209.7069 >387 178.2220 157.87687 198.5671 147.106804 209.3372 >388 170.4287 143.80863 197.0487 129.716832 211.1405 >389 162.6353 128.61257 196.6581 110.602006 214.6687 >390 154.8420 112.52646 197.1576 90.125956 219.5581 >391 147.0487 95.66491 198.4324 68.463984 225.6334 >392 139.2553 78.10706 200.4036 45.737114 232.7736 >393 131.4620 59.92462 202.9994 22.055013 240.8690 >394 123.6687 41.14798 206.1894 -2.535833 249.8732 >395 115.8753 21.82457 209.9261 -27.962900 259.7136 >'data.frame': 10 obs. of 5 variables: > $ Point Forecast: num 186 178 170 163 155 ... > $ Lo 80 : num 171 158 144 129 113 ... > $ Hi 80 : num 202 199 197 197 197 ... > $ Lo 95 : num 162.3 147.1 129.7 110.6 90.1 ... > $ Hi 95 : num 210 209 211 215 220 ... > >any idea on how to accomplish this? > >Best regards, > >Paul > > [[alternative HTML version deleted]] > >______________________________________________ >R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see >https://stat.ethz.ch/mailman/listinfo/r-help >PLEASE do read the posting guide >http://www.R-project.org/posting-guide.html >and provide commented, minimal, self-contained, reproducible code.-- Sent from my phone. Please excuse my brevity.