Send output to:
Browser Blue - Charts White
Browser Black/White
CSV
Data:
133448.00 132951.00 132447.00 131404.00 141722.00 141176.00 133448.00 128310.00 128807.00 128807.00 129360.00 130354.00 131901.00 131901.00 130907.00 128310.00 141722.00 143766.00 140679.00 133448.00 136542.00 131901.00 133994.00 134995.00 136038.00 133448.00 133994.00 130354.00 141722.00 145313.00 142226.00 136542.00 142723.00 136038.00 142226.00 141722.00 143269.00 137585.00 143766.00 143269.00 152544.00 150451.00 142226.00 138082.00 143766.00 136038.00 141722.00 142723.00 144816.00 140182.00 142723.00 144270.00 149954.00 145313.00 139132.00 132447.00 138635.00 121625.00 129857.00 134491.00 139132.00 132447.00 132447.00 132447.00 136038.00 130907.00 124173.00 118538.00 122626.00 106666.00 116445.00 122129.00 123172.00 117488.00 117985.00 116445.00 121625.00 117985.00 110810.00 105623.00 114394.00 95347.00 107716.00 113351.00 113351.00 106666.00 100485.00 99988.00 105623.00 100485.00 90713.00 83979.00 91210.00 74207.00 89663.00 97888.00 100485.00 94801.00 87619.00 92757.00 94801.00 93254.00 77791.00 70616.00 75747.00 60291.00 76251.00 81935.00 86569.00 78841.00 71610.00 75747.00 77791.00 73703.00 58247.00 51513.00 57694.00 40691.00 59241.00 70616.00
Seasonal period
0
12
1
2
3
4
5
6
7
8
9
10
11
12
Type of Exponential Smoothing
(?)
no
Single
Double
Triple
Type of seasonality
(?)
512
additive
multiplicative
Number of Forecasts
12
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Chart options
R Code
par1 <- as.numeric(par1) if (par2 == 'Single') K <- 1 if (par2 == 'Double') K <- 2 if (par2 == 'Triple') K <- par1 nx <- length(x) nxmK <- nx - K x <- ts(x, frequency = par1) if (par2 == 'Single') fit <- HoltWinters(x, gamma=F, beta=F) if (par2 == 'Double') fit <- HoltWinters(x, gamma=F) if (par2 == 'Triple') fit <- HoltWinters(x, seasonal=par3) fit myresid <- x - fit$fitted[,'xhat'] bitmap(file='test1.png') op <- par(mfrow=c(2,1)) plot(fit,ylab='Observed (black) / Fitted (red)',main='Interpolation Fit of Exponential Smoothing') plot(myresid,ylab='Residuals',main='Interpolation Prediction Errors') par(op) dev.off() bitmap(file='test2.png') p <- predict(fit, par1, prediction.interval=TRUE) np <- length(p[,1]) plot(fit,p,ylab='Observed (black) / Fitted (red)',main='Extrapolation Fit of Exponential Smoothing') dev.off() bitmap(file='test3.png') op <- par(mfrow = c(2,2)) acf(as.numeric(myresid),lag.max = nx/2,main='Residual ACF') spectrum(myresid,main='Residals Periodogram') cpgram(myresid,main='Residal Cumulative Periodogram') qqnorm(myresid,main='Residual Normal QQ Plot') qqline(myresid) par(op) dev.off() load(file='createtable') a<-table.start() a<-table.row.start(a) a<-table.element(a,'Estimated Parameters of Exponential Smoothing',2,TRUE) a<-table.row.end(a) a<-table.row.start(a) a<-table.element(a,'Parameter',header=TRUE) a<-table.element(a,'Value',header=TRUE) a<-table.row.end(a) a<-table.row.start(a) a<-table.element(a,'alpha',header=TRUE) a<-table.element(a,fit$alpha) a<-table.row.end(a) a<-table.row.start(a) a<-table.element(a,'beta',header=TRUE) a<-table.element(a,fit$beta) a<-table.row.end(a) a<-table.row.start(a) a<-table.element(a,'gamma',header=TRUE) a<-table.element(a,fit$gamma) a<-table.row.end(a) a<-table.end(a) table.save(a,file='mytable.tab') a<-table.start() a<-table.row.start(a) a<-table.element(a,'Interpolation Forecasts of Exponential Smoothing',4,TRUE) a<-table.row.end(a) a<-table.row.start(a) a<-table.element(a,'t',header=TRUE) a<-table.element(a,'Observed',header=TRUE) a<-table.element(a,'Fitted',header=TRUE) a<-table.element(a,'Residuals',header=TRUE) a<-table.row.end(a) for (i in 1:nxmK) { a<-table.row.start(a) a<-table.element(a,i+K,header=TRUE) a<-table.element(a,x[i+K]) a<-table.element(a,fit$fitted[i,'xhat']) a<-table.element(a,myresid[i]) a<-table.row.end(a) } a<-table.end(a) table.save(a,file='mytable1.tab') a<-table.start() a<-table.row.start(a) a<-table.element(a,'Extrapolation Forecasts of Exponential Smoothing',4,TRUE) a<-table.row.end(a) a<-table.row.start(a) a<-table.element(a,'t',header=TRUE) a<-table.element(a,'Forecast',header=TRUE) a<-table.element(a,'95% Lower Bound',header=TRUE) a<-table.element(a,'95% Upper Bound',header=TRUE) a<-table.row.end(a) for (i in 1:np) { a<-table.row.start(a) a<-table.element(a,nx+i,header=TRUE) a<-table.element(a,p[i,'fit']) a<-table.element(a,p[i,'lwr']) a<-table.element(a,p[i,'upr']) a<-table.row.end(a) } a<-table.end(a) table.save(a,file='mytable2.tab')
Compute
Summary of computational transaction
Raw Input
view raw input (R code)
Raw Output
view raw output of R engine
Computing time
0 seconds
R Server
Big Analytics Cloud Computing Center
Click here to blog (archive) this computation